Sie sind auf Seite 1von 299

Diabetes Mellitus in

Developing Countries
and Underserved

Sam Dagogo-Jack

Diabetes Mellitus in Developing
Countries and Underserved
Sam Dagogo-Jack

Diabetes Mellitus in
Developing Countries
and Underserved
Sam Dagogo-Jack
University of Tennessee Health
Science Center
Memphis, Tennessee

ISBN 978-3-319-41557-4 ISBN 978-3-319-41559-8 (eBook)

DOI 10.1007/978-3-319-41559-8

Library of Congress Control Number: 2016959999

Springer International Publishing Switzerland 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, express or implied, with respect to the material
contained herein or for any errors or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG Switzerland
The registered company address is Gewerbestrasse 11, 6330 Cham, Switzerland
This volume is dedicated to all persons who struggle with the
syndrome of diabetes mellitus, the global community of
clinicians who manage diabetes, and the researchers dedicated
to finding better methods for the treatment, prevention, and,
ultimately, cure of diabetes mellitus.

No other disorder exerts a more pervasive and catholic global burden on

humanity and society than does diabetes. The current diabetes epidemic is
felt on all continents, by all ethnic groups, among the rich and poor, and
across gender and age groups. In the first Global Diabetes Report released in
2016, the World Health Organization noted that diabetes is no longer a dis-
ease of predominantly rich nations and that the prevalence of diabetes is
steadily increasing everywhere, most markedly in the worlds middle-income
countries. There were an estimated 108 million adults with diabetes globally
in 1980. By 2014, that number had risen to 422 million adults, and the global
estimate is projected to increase to 592 million adults with diabetes by 2035.
Sadly, low- and middle-income countries are projected to experience the
steepest increase as they transition to lower-middle and upper-middle income
status. The complications of diabetes, such as blindness, kidney failure,
amputations, heart disease, stroke, and mortality, are a personal tragedy as
well as a threat to national and global security, through their negative effects
on the vitality and economic productivity of afflicted citizens.
Low- and middle-income countries have limited resources for dealing
with the costs of controlling diabetes and its complications. Even in the
developed economies, the quality of diabetes control is suboptimal for
approximately 50 % of the affected population. Moreover, socioeconomically
vulnerable demographic subgroups have always existed in North America,
Europe, Australia, and New Zealand; individuals from those groups often
experience disparities in diabetes prevalence, quality of care, and outcomes.
Autochthonous or aboriginal people and migrant groups predominate among
the vulnerable and often underserved populations with regard to diabetes.
The chronic low-grade human migration trends (usually from lower-income
to upper-income regions) have escalated exponentially as a result of recent
geopolitical upheavals. The flood of new immigrants to Europe and elsewhere
has added to the challenges of providing optimal diabetes care in a setting of
rapid cultural, linguistic, and socioeconomic metamorphosis.
Diabetes Mellitus in Developing Countries and Underserved Communities,
a global textbook of diabetes, presents a detailed consideration of the epide-
miology, genomic landscape, clinical presentation, complications, manage-
ment, and prevention of diabetes in all regions of the world. The regions
covered include sub-Saharan Africa, the Middle East and North Africa,
Eastern Europe, China and the Western Pacific, India and Southeast Asia,
Latin America and the Caribbean, as well as Australia, New Zealand, North

viii Preface

America, and Western Europe. In the latter more affluent regions, appropriate
focus is placed on vulnerable indigenous and immigrant populations.
Additional attention is given to a description of regional coordination of dia-
betes care, diabetes prevention services, preferences and imperatives regard-
ing pharmacotherapy, and the status of diabetes research. The chapters
conclude with the identification of future directions, unmet needs, unan-
swered questions, or even unquestioned answers, with regard to diabetes in
specific regions.
The availability, in a single volume, of comprehensive information on the
peculiarities of diabetes pathophysiology, genetics, and best practices in
major regions of the world, crafted by leading authorities in the field, is a
valuable resource for clinicians, researchers, scholars, public health leaders,
students, and everyone interested in diabetes. The authors are to be com-
mended for their hard work in creating this informed and informative treatise
on global diabetes.
The conceptual insight of Kristopher Spring (editor, Clinical Medicine)
and logistical support from Ms. Saanthi Shankhararaman (project coordina-
tor), both at Springer, greatly facilitated the creation of this book. Much grati-
tude is owed to them for their professionalism and engagement that resulted
in the successful execution of the book project.

Memphis, TN, USA Sam Dagogo-Jack


1 The Global Burden of Diabetes: An Overview . . . . . . . . . . . . . . . . 1

William H. Herman
2 Primary Prevention of Type 2 Diabetes: An Imperative
for Developing Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Sam Dagogo-Jack
3 Diabetes in Sub-Saharan Africa . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Felix Assah and Jean Claude Mbanya
4 Type 2 Diabetes in the Middle East and
North Africa (MENA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Yasmin Khan and Osama Hamdy
5 Diabetes in China and the Western Pacific Region . . . . . . . . . . . 63
Juliana C.N. Chan, Elaine Y.K. Chow, and
Andrea A.O. Luk
6 Diabetes in India and Southeast Asia . . . . . . . . . . . . . . . . . . . . . . 85
Shashank R. Joshi and S.R. Aravind
7 Diabetes in Latin America . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Omar Y. Bello-Chavolla and Carlos A. Aguilar-Salinas
8 Diabetes in the Caribbean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Michael S. Boyne
9 Diabetes in Indigenous Australians and Other
Underserved Communities in Australia . . . . . . . . . . . . . . . . . . . 151
Stephen Colagiuri
10 Diabetes Among Mori and Other Ethnic Groups
in New Zealand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Evan Atlantis, Grace Joshy, Margaret Williams,
and David Simmons
11 Diabetes in Eastern Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Mykolay Khalangot, Vitaliy Gurianov,
Alexander Vaiserman, Ieva Strele, Vasile Fedash,
and Victor Kravchenko

x Contents

12 Diabetes in Ethnic Minorities and Immigrant

Populations in Western Europe . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Oliver Razum and Helmut Steinberg
13 Diabetes Among Indigenous Canadians . . . . . . . . . . . . . . . . . . . 235
Sudaba Mansuri and Anthony J. Hanley
14 Diabetes in Native Populations and Underserved
Communities in the USA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
Joshua J. Joseph and Sherita Hill Golden
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285

S.R. Aravind, MBBS, DNB, FRCP, FDRC, DHA, FRSSDI Columbia

Asia Hospital, Yeshwanthpur, Bangalore, India
Felix Assah, MD, PhD Faculty of Medicine and Biomedical Sciences,
University of Yaound I, Yaound, Cameroon
Evan Atlantis, PhD Western Sydney University, Campbelltown, NSW,
Omar Y. Bello-Chavolla Department of Endocrinology and Metabolism,
Instituto Nacional de Ciencias Mdicas y Nutricin Salvador Zubirn,
Mexico City, Mexico
Facultad de Medicina, Universidad Nacional Autnoma de Mxico, Mexico
City, Mexico
Michael S. Boyne, MD, FRCPC Tropical Metabolism Research Unit,
Tropical Medicine Research Institute, The University of the West Indies,
Mona, Kingston, Jamaica
Juliana C.N. Chan, MD, FRCP The Chinese University of Hong Kong,
Prince of Wales Hospital, Shatin, Hong Kong, China
Elaine Y.K. Chow, MBChB, PhD Department of Medicine and Therapeutics,
Hong Kong Institute of Diabetes, Hong Kong, SAR, China
Obesity and Li Ka Shing Institute of Health Sciences, The Chinese University
of Hong Kong, Prince of Wales Hospital, Hong Kong, SAR, China
Stephen Colagiuri, MD W.H.O. Collaborating Centre on Physical Activity,
Nutrition and Obesity, The University of Sydney, Sydney, NSW, Australia
Sam Dagogo-Jack, MD, MBBS, MSc, FRCP, FACP, FACE Division of
Endocrinology, Diabetes & Metabolism, University of Tennessee Health
Science Center, Memphis, TN, USA
Vasile Fedash, MD, PhD Institute of Physiology and Sanocreatology of
ASM, Chisinau, Moldova
Sherita Hill Golden, MD, MHS Department of Medicine, Johns Hopkins
University School of Medicine, Baltimore, MD, USA

xii Contributors

Vitaliy Gurianov, PhD Bogomolets National Medical University, Kyiv,

Osama Hamdy, MD, PhD, FACE Joslin Diabetes Center, Harvard Medical
School, Boston, MA, USA
Anthony J. Hanley, PhD Departments of Nutritional Sciences and Public
Health Sciences, Faculty of Medicine, University of Toronto, Toronto, ON,
William H. Herman, MD, MPH University of Michigan, Ann Arbor, MI,
Joshua J. Joseph, MD Division of Endocrinology, Diabetes and Metabolism,
Johns Hopkins University School of Medicine, Baltimore, MD, USA
Shashank R. Joshi, MD, DM, FICP, FACP, FACE, FRCP Joshi Clinic,
Lilavati & Bhatia Hospital, Mumbai, India
Grace Joshy, PhD Australian National University, Canberra, ACT, Australia
Mykolay Khalangot, MD, Sc D Shupyk National Academy of Postgraduate
Medical Education, Kyiv, Ukraine
Yasmin Khan, MD, MPH Adult Diabetes Section, Joslin Diabetes Center,
Boston, MA, USA
Victor Kravchenko, Sc D Komisarenko Institute of Endocrinology and
Metabolism, National Academy of Medical Sciences, Kyiv, Ukraine
Andrea A.O. Luk, MBBS, MRCP Department of Medicine and Therapeutics,
Hong Kong Institute of Diabetes, Hong Kong, SAR, China
Obesity and Li Ka Shing Institute of Health Sciences, The Chinese University
of Hong Kong, Prince of Wales Hospital, Hong Kong, SAR, China
Sudaba Mansuri Department of Nutritional Sciences, Faculty of Medicine,
University of Toronto, Toronto, ON, Canada
Jean Claude Mbanya, MD, PhD, FRCP Faculty of Medicine and Biomedical
Sciences, University of Yaounde I, Yaounde, Cameroon
Oliver Razum Fakultt frGesundheitswissenschaften, Universitt Bielefeld,
School of Public Health, Bielefeld, Germany
Carlos A. Aguilar Salinas Departamento de Endocrinologia y Metabolismo,
Instituto Nacional de CienciasMdicas y Nutricion, Mexico City, Mexico
Christopher D. Saudek, MD Division of Endocrinology, Diabetes and
Metabolism, Johns Hopkins University School of Medicine, Baltimore, MD,
David Simmons, MBBS, MA, FRACP, FRCP, MD Western Sydney University,
Campbelltown, NSW, Australia
Contributors xiii

Helmut Steinberg, MD Department of Medicine, Division of Endocrinology,

Diabetes and Metabolism, University of Tennessee Health Science Center,
Memphis, TN, USA
Ieva Strele, MD, PhD Riga Stradins University, Riga, Latvia
Alexander Vaiserman, Sc D Chebotariov Institute of Gerontology, National
Academy of Medical Sciences, Kyiv, Ukraine
Margaret Williams, PhD Auckland University of Technology, Auckland,
New Zealand
About the Editor

Sam Dagogo-Jack, MD, is Professor of

Medicine and Chief of the Division of
Endocrinology, Diabetes and Metabolism at
the University of Tennessee Health Science
Center, Memphis, Tennessee, where he holds
the A. C. Mullins Endowed Chair in
Translational Research. He is also Director of
the Endocrinology Fellowship Training
Program and Director of the General Clinical
Research Center at UTHSC. A global citizen,
Dr. Dagogo-Jack has studied, practiced,
researched, and/or taught medicine and dia-
betology in Africa, Asia, Australia, Europe, North and South America,
Caribbean countries, and the Middle East. His current research focuses on the
interaction of genetic and environmental factors in the prediction and preven-
tion of prediabetes and diabetes. Dr. Dagogo-Jack is a recipient of the Banting
Medal for Leadership from the American Diabetes Association and the
Distinction in Endocrinology Award from the American College of
Endocrinology. He served as the 2015 President, Medicine & Science, of the
American Diabetes Association.

The Global Burden of Diabetes:
An Overview 1
William H. Herman

Introduction [5]. Whatever the causes, the estimates are con-

servative, and the future burden of diabetes is
The global burden of diabetes is enormous and likely to be even greater than projected.
growing. Between 1997 and 2010, four research In 2015, the International Diabetes Federation
groups estimated the numbers of people with dia- (IDF) estimated that globally, 415 million adults
betes globally and projected the future burden of 2079 years of age or 8.8 % of the adult popula-
diabetes (Fig. 1.1) [14]. In each instance, the pre- tion have diabetes [5]. There are 321 million
vious estimate was shown to have underestimated people of working age (2064 years) with diabe-
the number of people with diabetes, and the pro- tes and 94 million people aged 6579 with diabe-
jections painted an ever more alarming picture of tes. There are slightly more men than women
the future burden of diabetes. The latest study with diabetes (215 million men vs. 200 million
estimated that between 2010 and 2030, the num- women). Currently, there are more people with
ber of adults with diabetes worldwide would diabetes in urban areas (270 million) than in rural
increase by 54 %, from 285 million to 439 million areas (145 million). Age-adjusted diabetes preva-
[4]. Much of the growth in the numbers of people lence rates among adults 2079 years of age vary
with diabetes would occur in middle-aged work- by region and by country. In 2015, the age-
ing adults [4]. The number of adults with diabetes adjusted diabetes prevalence was 11.5 % in North
would increase by 20 % in developed countries America and the Caribbean, 10.7 % in the Middle
and by 69 % in developing countries [4]. A part of East and North Africa, 9.6 % in South and Central
the increase in the number of people with diabetes America, 8.8 % in the Western Pacific and
worldwide is due to increased incidence of type 2 Southeast Asia, 7.3 % in Europe, and 3.8 % in
diabetes related to urbanization, obesogenic diets, Africa [5]. Perhaps most alarmingly, approxi-
and decreased physical activity [5]. A larger part mately 75 % of people with diabetes live in low-
of the increase is due simply to population growth, and middle-income countries [5].
aging of the population, and decreased mortality Diabetes prevalence rates vary greatly among
indigenous communities. These communities are
characterized by unique languages, knowledge
systems, and beliefs [5]. Many have a special
W.H. Herman, MD, MPH
Departments of Internal Medicine and Epidemiology, relationship with their traditional land which
University of Michigan, often has a fundamental importance for their cul-
1000 Wall Street, Room 6108/SPC 5714, Ann Arbor, ture [5]. In many cases, the prevalence of diabe-
MI 48105-1912, USA tes is greater in indigenous populations than in

Springer International Publishing Switzerland 2017 1

S. Dagogo-Jack (ed.), Diabetes Mellitus in Developing Countries and Underserved Communities,
DOI 10.1007/978-3-319-41559-8_1
2 W.H. Herman

350 Amos, McCarty & Zimmet, 1997


King, Aubert & Herman, 1998

200 Wild, Roglic, Green, Sicree & King,
Shaw, Sicree & Zimmet, 2010
1990 2000 2010 2020 2030 2040

Fig. 1.1 Global estimates and projections of the number of people with diabetes

the surrounding population such as in indige- in diabetes prevalence will occur in regions where
nous Australian [6], New Zealand Mori [7], economies are moving from low-income to mid-
Greenland Inuit [8], Canadian Inuit [9], and dle-income levels. By 2040, the IDF has estimated
Native American populations [10]. Some indig- that the age-adjusted prevalence of diabetes will
enous populations that still live very traditional be 12.0 % in North America, 9.7 % in South and
lifestyles have a relatively low prevalence of dia- Central America, 9.0 % in the Western Pacific,
betes [5]. 9.1 % in Southeast Asia, 7.6 % in Europe, and
Not every person with diabetes has been diag- 4.2 % in Africa [5].
nosed. Globally, it is estimated that one-in-two The global epidemic of type 2 diabetes has
adults with diabetes is undiagnosed [5]. In major implications for healthcare expenditures.
regions where healthcare resources are lacking Most countries dedicate between 5 % and 20 % of
such as in sub-Saharan Africa, the proportion of their total healthcare resources to treat diabetes
people with diabetes who are undiagnosed is as and its complications [5]. In 2015, the IDF esti-
high as 67 % [5]. Even in high-income regions mated that total diabetes healthcare expenditures
such as North America, slightly more than one- for persons 2079 years of age were $673 billion
third of people with diabetes have not been diag- or 12 % of total healthcare expenditures world-
nosed [5]. Globally, 81 % of people with diabetes wide [5]. The IDF has projected that this number
who are undiagnosed live in low- and middle- will increase to $802 billion by 2040 assuming
income countries [5]. constant per capita healthcare expenditures [5].
The IDF has estimated that by 2040, 642 mil- The increase in healthcare expenditures for diabe-
lion adults 2079 years of age or 10.4 % of the tes by 2040 is projected to be proportionately
world population will have diabetes [5]. In 2040, less than the increase in the number of people
proportionately more people with diabetes will be with diabetes because countries with the largest
6579 years of age. The gender difference increases in the numbers of people with diabetes
between men and women is expected to decrease are those with the lowest per capita spending for
modestly by 2040, but the global difference in the diabetes. In 2015, an average of $1,622 was spent
number of people with diabetes in urban and rural per person with diabetes [5]. The range in expen-
areas is expected to widen with 478 million peo- ditures for diabetes varied by almost 85-fold:
ple with diabetes living in urban areas and 164 from $7,859 per person per year in North America
million living in rural areas. The largest increases and the Caribbean, $2,610 in Europe, $1,169 in
1 The Global Burden of Diabetes: An Overview 3

South and Central America, $693 in the Western increased demand for and access to expensive but
Pacific, $483 in the Middle East and North Africa, lifesaving care. As low- and middle-income coun-
$243 in Africa, and $93 in Southeast Asia [5]. tries develop economically, the demand for and
Not surprisingly, these dramatic differences in per capita expenditures for the treatment of diabe-
mean annual healthcare expenditures per person tes and its complications will increase substan-
with diabetes are associated with major differ- tially. The future costs of healthcare for diabetes
ences in the types of expenditures for diabetes. In are likely to be many times higher than projected
developed countries where mean annual health- by the IDF and, if costs remain unchecked, have
care expenditures are high, a large absolute the potential to bankrupt healthcare systems and
amount but a small proportion of total healthcare indeed national economies.
expenditures are for antihyperglycemic therapy What can be done? All countries, but especially
[11]. The greatest proportion of healthcare expen- low- and middle-income countries with limited
ditures for diabetes go for the treatment of com- resources, must carefully balance the benefits and
plications and comorbidities. In contrast, in costs of allocating scarce healthcare resources for
developing countries where mean annual health- population-level and targeted interventions for
care expenditures are low, the greatest proportion diabetes prevention, for antihyperglycemic and
of expenditures go for antihyperglycemic therapy cardiovascular therapies to delay or prevent the
[11]. Although some resources go to the treat- development of complications among people with
ment of acute metabolic and infectious complica- diabetes, and for the treatment of advanced dia-
tions, little is spent on the treatment of chronic betic complications. Addressing only the health-
complications and comorbidities such as renal care needs of individuals with diagnosed diabetes
and cardiovascular disease. will leave the epidemic of type 2 diabetes and its
Economic development is associated with downstream complications, comorbidities, and
increased per capita healthcare expenditures. In a costs unchecked. Clearly, tailored strategies that
recent study, Seuring and colleagues examined allocate resources to prevent diabetes, to treat dia-
the factors associated with the heterogeneity in betes, and to treat its complications and comor-
diabetes healthcare expenditures among coun- bidities will be required.
tries [12]. Their work demonstrated that the direct Population-level policy interventions address-
costs of diabetes are positively associated with a ing food supply, the built environment, tax pol-
countrys per capita gross domestic product icy, and financial incentives and disincentives all
(GDP). Healthcare expenditures increase with offer great promise in addressing the obesogenic
national economic wealth. Per capita GDP and diabetogenic environment at a relatively low
explained about one-third of the variation in dia- cost. Examples abound in the area of tobacco
betes healthcare expenditures among countries control [14]. Unfortunately, there are many dif-
such that every additional dollar in per capita ferences between smoking, diet, and physical
GDP translated into an average increase in direct activity and the evidence base for population-
diabetes expenditures of about $0.04 [12]. level interventions for diabetes prevention is
This phenomenon is further illustrated by a limited.
study that assessed rates of end-stage renal dis- The evidence base for targeted interventions
ease (ESRD) treatment by national wealth [13]. In to delay or prevent the development of type 2 dia-
low- and middle-income countries with per capita betes is extensive, robust, and consistent. At least
GDP less than ~$10,000, rates of ESRD treatment four trials from China [15], Finland [16], the
were very low but tended to increase with per United States [17], and India [18] have demon-
capita GDP. In contrast, in high-income countries strated that lifestyle interventions that achieve a
with per capita GDP above ~$10,000, rates of 57 % reduction in initial body weight and
ESRD treatment were consistently higher, reflect- increase brisk walking to approximately 150 min
ing greater access to ESRD treatment. Economic per week can reduce the incidence of type 2 dia-
development thus appeared to be associated with betes by 2958 % in high-risk individuals with
4 W.H. Herman

impaired glucose tolerance. Other trials have change from glucose to HbA1c possible in all popula-
tions? J Clin Endocrinol Metab. 2010;95:E3336.
demonstrated the efficacy of metformin [17, 18],
9. Diabetes in Canada: facts and figures from a public
alpha-glucosidase inhibitors [19, 20], and thia- health perspective. Ottawa Ont: Public Health Agency
zolidinediones [2123] for diabetes prevention. of Canada; 2011.
Unfortunately, such interventions required sub- 10. Lee ET, Howard BV, Savage PJ, Cowan LD, Fabsitz
RR, Oopik AJ, Yeh J, Go O, Robbins DC, Welty
stantial resources to identify at-risk individuals,
TK. Diabetes and impaired glucose tolerance in three
to implement interventions, and to maintain long- American Indian populations aged 4574 years. The
term adherence. In addition, the evidence that Strong Heart Study. Diabetes Care. 1995;18:599610.
targeted interventions can be translated into long- 11. Zhang P, Zhang X, Brown J, Vistisen D, Sicree R,
Shaw J, Nichols G. Global healthcare expenditure on
term clinical practice in large, less highly selected
diabetes for 2010 and 2030. Diabetes Res Clin Pract.
populations is lacking. 2010;87:293301.
The challenge is clear. As reported by the IDF, 12. Seuring T, Archangelidi O, Suhrcke M. The economic
Type 2 diabetes is a global epidemic with devas- costs of type 2 diabetes: a global systematic review.
Pharmacoeconomics. 2015;33:81131.
tating humanitarian, social, and economic conse-
13. Grassmann A, Gioberge S, Moeller S, Brown G. End-
quences [5]. We must understand the global stage renal disease: global demographics in 2005 and
burden of diabetes to increase awareness, to plan observed trends. Artif Organs. 2006;30:8957.
for future needs, and to inform interventions. An 14. Reducing risks and preventing disease: population-
wide interventions. In: World Health Organization
understanding of the present and future burden of
(editor). Global status report on noncommunicable
diabetes in developing countries and underserved diseases 2010. Chapter 4. Geneva; 2011. http://www.
communities is essential to this task. .
Accessed 5 Jan 2016.
15. Pan XR, Li GW, Hu YH, et al. Effects of diet and
exercise in preventing NIDDM in people with
References impaired glucose tolerance. The Da Qing IGT and
Diabetes Study. Diabetes Care. 1997;20:53744.
1. Amos AF, McCarty DJ, Zimmet P. The rising global 16. Tuomilehto J, Lindstrm J, Eriksson JG, Finnish
burden of diabetes and its complications: estimates Diabetes Prevention Study Group, et al. Prevention of
and projections to the year 2010. Diabet Med. 1997;14 type 2 diabetes mellitus by changes in lifestyle among
Suppl 5:S185. subjects with impaired glucose tolerance. N Engl
2. King H, Aubert RE, Herman WH. Global burden of J Med. 2001;344:134350.
diabetes, 19952025: prevalence, numerical estimates 17. Knowler WC, Barrett-Connor E, Fowler SE, Diabetes
and projections. Diabetes Care. 1998;21:141431. Prevention Program Research Group, et al. Reduction in
3. Wild S, Roglic G, Green A, Sicree R, King H. Global the incidence of type 2 diabetes with lifestyle interven-
prevalence of diabetes: estimates for the year 2000 tion or metformin. N Engl J Med. 2002;346:393403.
and projections for 2030. Diabetes Care. 18. Ramachandran A, Snehalatha C, Mary S, Mukesh B,
2004;27:104753. Bhaskar AD, Vijay V, Indian Diabetes Prevention
4. Shaw JE, Sicree RA, Zimmet PZ. Global estimates of Programme (IDPP). The Indian Diabetes Prevention
the prevalence of diabetes for 2010 and 2030. Diabetes Programme shows that lifestyle modification and met-
Res Clin Pract. 2010;87:414. formin prevent type 2 diabetes in Asian Indian sub-
5. International Diabetes Federation. IDF diabetes atlas. jects with impaired glucose tolerance (IDPP-1).
7th ed. Brussels: International Diabetes Federation; Diabetologia. 2006;49:28997.
2015. Accessed 5 Jan 19. Chiasson JL, Josse RG, Gomis R, Hanefeld M,
2015. Karasik A, Laakso M, STOP-NIDDM Trail Research
6. Minges KE, Zimmet P, Magliano DJ, Dunstan DW, Group. Acarbose for prevention of type 2 diabetes
Brown A, Shaw JE. Diabetes prevalence and determi- mellitus: the STOP-NIDDM randomised trial. Lancet.
nants in Indigenous Australian populations: a system- 2002;359:20727.
atic review. Diabetes Res Clin Pract. 2011;93:13949. 20. Kawamori R, Tajima N, Iwamoto Y, Kashiwagi A,
7. Coppell KJ, Mann JI, Williams SM, Jo E, Drury PL, Shimamoto K, Kaku K, Voglibose Ph-3 Study Group.
Miller J, et al. Prevalence of diagnosed and undiag- Voglibose for prevention of type 2 diabetes mellitus: a
nosed diabetes and prediabetes in New Zealand: find- randomised, double-blind trial in Japanese individuals
ings from the 2008/09 Adult Nutrition Survey. N Z with impaired glucose tolerance. Lancet. 2009;373:
Med J. 2013;126:2342. 160714.
8. Jrgensen ME, Bjerregaard P, Borch-Johnsen K, 21. Buchanan TA, Xiang AH, Peters RK, et al.
Witte D. New diagnostic criteria for diabetes: is the Preservation of pancreatic beta-cell function and pre-
1 The Global Burden of Diabetes: An Overview 5

vention of type 2 diabetes by pharmacological treat- fasting glucose: a randomised controlled trial. Lancet.
ment of insulin resistance in high-risk hispanic 2006;368:1096105. Erratum in: Lancet 2006;
women. Diabetes. 2002;51:2796803. 368:1770.
22. DREAM (Diabetes REduction Assessment with 23. DeFronzo RA, Tripathy D, Schwenke DC, ACT
ramipril and rosiglitazone Medication) Trial NOW Study, et al. Pioglitazone for diabetes preven-
Investigators, Gerstein HC, Yusuf S, Bosch J, et al. tion in impaired glucose tolerance. N Engl J Med.
Effect of rosiglitazone on the frequency of diabetes in 2011;364:110415. Erratum in: N Engl J Med
patients with impaired glucose tolerance or impaired 2011;365:189. N Engl J Med 2011;365:869.
Primary Prevention of Type 2
Diabetes: An Imperative 2
for Developing Countries

Sam Dagogo-Jack

Introduction again with the fastest increases occurring in

developing countries. Although sub-Saharan
In 2015, the International Diabetes Federation Africa (SSA) currently has the lowest estimate
(IDF) estimated that 415 million adults had dia- for diabetes prevalence, the steepest proportion-
betes worldwide (1). Type 2 diabetes mellitus ate increase in diabetes (240 %) is projected to
(T2DM) accounts for 9095 % of all diabetes occur in that region [1] (Fig. 2.1). In contrast,
cases. The 2015 IDF estimates for people with recent data indicate that the incidence of diabetes
diabetes in regions of the world (descending is leveling off in the United States of America,
order) were 153.2 million in the Western Pacific, presumably as a result of increased awareness
59.8 million in Europe, 44.3 million in North and national diabetes prevention initiatives [2].
America and the Caribbean, 35.3 million in the Furthermore, major declines in the rates of diabe-
Middle East/North Africa, and 14.2 million in tes complications have also occurred in the
sub-Saharan Africa [1]. The country-specific United States [3]. Unfortunately, these encourag-
estimates showed that the top 10 nations with the ing developments in the United States are not
highest numbers of people with diabetes are shared by most regions of the world, where the
China (109.6 million), India (99.2 million), the escalating prevalence of diabetes and ravages
United States of America (29.3 million), Brazil from the disease persist inexorably [1, 46].
(14.3 million), the Russian Federation (12.1 mil-
lion), Mexico (11.5 million), Indonesia (10 mil-
lion), Egypt (7.8 million), Japan (7.2 million), Risk Factors for Type 2 Diabetes
and Bangladesh (7.1 million) [1]. Thus, countries
in the developing economies contribute dispro- Both genetic and environmental factors underlie
portionately to the escalating global diabetes bur- the development of T2DM [6, 7]. To date, more
den (13). By 2040, the number of adults with than 60 gene variants associated with T2DM
diabetes is projected to increase to 642 million, have been identified; however, the effect size of
these individual gene variants is rather modest
[812]. The environmental risk factors strongly
associated with T2DM risk include obesity, phys-
S. Dagogo-Jack, MD, MBBS, MSc, FRCP,
FACP, FACE ical inactivity, history of gestational diabetes,
Division of Endocrinology, Diabetes & Metabolism, hypertension, and dyslipidemia, among others
University of Tennessee Health Science Center, (Table 2.1) [6, 7, 13]. These risk factors interact
920 Madison Avenue, Memphis, TN 38163, USA with genetic predisposition (indicated by a family

Springer International Publishing Switzerland 2017 7

S. Dagogo-Jack (ed.), Diabetes Mellitus in Developing Countries and Underserved Communities,
DOI 10.1007/978-3-319-41559-8_2
8 S. Dagogo-Jack

Fig. 2.1 Projected

increases in diabetes
prevalence in different
regions of the world, 642
20152040 (Source:
International Diabetes million
Federation [1]) people
living with

415 Africa 141%

million Middle East & North Africa 104%

South East Asia 82%

South and Central America 65%

Western Pacific 40%

Europe 19%

North America & Caribbean 37%

2015 2040

history of diabetes and/or high-risk ethnic heri- Table 2.1 Risk factors for type 2 diabetes
tage) to promote the development of diabetes. Physical inactivity
Such a phenomenon was clearly demonstrated in Overweight/obesity
the studies that showed a threefold increase in the First-degree relative with diabetes
rate of T2DM in recent Japanese immigrants to High-risk ethnicity
the United States compared with native Japanese Gestational diabetes or delivery of a baby weighing
[14]. Since a dramatic increase in disease preva- 9 lb or greater
lence over a relatively short time frame in humans HDL cholesterol <35 mg/dl TG >250 mg/dl
is unlikely to be due to sudden new genetic muta- Hypertension (>140/90 mmHg or on therapy)
tions, environmental factors (notably changes in A1C 5.7, IGT, or IFG on previous testing
diet, physical activity, and perhaps microbial Conditions associated with insulin resistance:
acanthosis nigricans, polycystic ovary disease, etc.
flora) probably trigger the surging diabetes rates
History of cardiovascular disease
among genetically predisposed populations. The
exact mechanisms whereby these environmental
triggers induce diabetes in genetically predis- resistance, impaired insulin secretion, impaired
posed persons remain to be fully elucidated. glucagon suppression, increased lipolysis, exag-
gerated hepatic glucose production, incretin defi-
ciency/resistance, maladaptive renal glucose
Pathophysiology reabsorption, and central nervous system defects
(including impaired dopaminergic tone and dys-
Current understanding indicates that multiple regulation of satiety) [1517] (Fig. 2.2). Insulin
pathophysiological defects underlie T2DM. resistance can be inherited or acquired. Obesity,
Generally, at least eight unique pathophysiological aging, physical inactivity, overeating, increased
defects are currently recognized in T2DM: insulin lipolysis, and accumulation of excessive amounts
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 9

Fig. 2.2 The major

pathophysiological Insulin
defects that lead to the resistance

in d

m se
development and se aire

pa ea
progression of type 2 cre d i

do ecr
tio nsu
diabetes. Many of the n lin

same defects (notably

Increased glucose
insulin resistance,
impaired insulin

secretion, lipolysis,

and subnormal incretin
response) at some
degree of expression
have also been
described in people
with prediabetes

cr go d
se ca ase
et n


gl cre
def tin


res iency

ista /
nce Increased hepatic
glucose production

of nonesterified (free) fatty acids are known causes the formation of intracellular DAG. Thus, multiple
of insulin resistance. Normally, cytoplasmic long- metabolic pathways link intracellular glucose
chain fatty acids are transported into mitochondria abundance (usually derived from carbohydrate
as long-chain fatty acyl coenzyme A (LCFA-CoA) consumption) to impaired fat oxidation, fatty acid
for beta-oxidation, a process that is gated by carni- synthesis, accumulation of long-chain fatty acids,
tine palmitoyl transferase (CPT)-1 and CPT-2 (the risk of lipotoxicity, and insulin resistance (Fig. 2.3).
shuttle enzymes located in the outer and inner Among the potent interventions that have been
mitochondrial membrane). This shuttle process demonstrated to ameliorate the pathological cellu-
ensures that fatty acids do not accumulate exces- lar and molecular processes leading to insulin
sively in the cytoplasm. Inhibition of that process resistance are caloric restriction (reduction of car-
leads to intracellular accumulation of long-chain bohydrate and fat intake), physical activity, and
fatty acids, which can induce lipotoxicity, cellular weight loss [1927].
dysfunction, and cell death [16, 17]. Further, intra-
cellular accumulation of long-chain fatty acids
along with diacylglycerol (DAG) can activate cer- Prediabetes
tain isoforms of protein kinase C (PKC), leading
to aberrant phosphorylation of the insulin receptor The term prediabetes refers to impaired glu-
and consequent insulin resistance. cose tolerance (IGT) and impaired fasting glu-
Acetyl-CoA, a product of glycolysis in the cose (IFG), two intermediate metabolic states
Krebs cycle, can be converted to malonyl CoA by between normal glucose tolerance and diabetes.
the enzyme acetyl-CoA carboxylase (ACC). IGT is defined by a plasma glucose level of
Malonyl-CoA is the activated two-carbon donor 140 mg/dl to 199 mg/dl (7.811.1 mmol/l), 2 h
required for fatty acid synthesis. Malonyl-CoA following ingestion of a 75-g oral solution. IFG is
also is a potent inhibitor of CPT-1, thereby block- defined by a fasting plasma glucose level of 100
ing the delivery and oxidation of fatty acids in the 125 mg/dl (5.66.9 mmol/l) [13] (Fig. 2.4). There
mitochondria. The result is accumulation of long- is considerable overlap in the risk factors and
chain fatty acids in the cytosol and eventual lipo- pathophysiological defects that underlie T2DM
toxicity [17, 18]. Glucose abundance also increases and prediabetes. Although the exact sequence of
10 S. Dagogo-Jack


Insulin receptor

G (-)
Malonyl CoA CPT
Acetyl CoA

Fig. 2.3 Schematic diagram of insulin signaling path- tylpalmitoyltransferase (CPT), the mitochondrial enzyme
ways and interactions with glucose (G) and fatty acid that transports long-chain fatty acids (LCFA) into the inner
metabolism. Increased intracellular glucose flux generates mitochondrial space for oxidation. The resultant accumu-
acetyl coenzyme A molecules which can be converted to lation of LCFA in the cytosol is linked to insulin resistance
malonyl coenzyme, a process catalyzed by acyl coenzyme via mechanisms involving protein kinase C (PKC) and
A carboxylase (ACC). Malonyl coenzyme A inhibits carni- altered phosphorylation of the insulin receptor

evolution of individual pathophysiological following features: (1) baseline fasting plasma

defects has not been determined precisely, many glucose (FPG) and the 2-h OGTT glucose values
of the defects coevolve during the pathogenesis are positively associated with diabetes risk; (2)
of T2DM and are demonstrable even at the stage the rate of progression from IGT to T2DM was
of prediabetes (Fig. 2.1) [2838]. Estimates by exponential among subjects in the top quartile of
the Centers for Disease Control and Prevention baseline FPG but increased linearly with increas-
(CDC) in the United States indicated that there ing 2-h OGTT glucose levels; (3) incident diabe-
were ~29 million adults with diabetes and 86 mil- tes occurred at higher rates in Hispanic,
lion with prediabetes in 2014 [39]. Worldwide, Mexican-Americans, Pima, and Nauruan popula-
there are more than 400 million people with pre- tions than among Caucasians; (4) the degree of
diabetes [1]. Individuals with prediabetes prog- obesity, as measured by the BMI, predicted
ress to T2DM at an annual rate of ~10 % [31, 32]. T2DM risk in three studies with the lowest inci-
dence rates of diabetes but not in the studies that
recorded the highest incidence of T2DM; and (5)
Predictors of Progression a family history of diabetes did not predict the
from Prediabetes to Type 2 risk of progression from IGT to diabetes in these
Diabetes studies, suggesting that genetic effects probably
are fully established by the stage of IGT [40].
An analysis of six prospective studies [40] on Thus, the magnitude of fasting and post-challenge
progression from IGT to diabetes revealed the dysglycemia (a reflection of insulin action and
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 11

Fig. 2.4 Criteria for

the definition of normal Criteria for the Diagnosis of Diabetes and Prediabetes
glucose regulation and
diagnosis of prediabetes
and diabetes. FPG Normal Prediabetes Diabetes
fasting plasma glucose,
2hPG two-hour
post-load plasma
glucose, OGTT oral
Fasting FPG < 100 mg/dl 100 125 mg/dl FPG 126 mg/dl
glucose tolerance test, Glucose (<5.6 mmol/l) (5.6 6.9 mmol/l) (7.0 mmol/l)
using standard 75 g

200 mg/dl (11.1

OGTT 2hPG <140mg/dl 140 199 mg/dl mmol/l) or random
(<7.8 mmol/l) (7.8 - 11.0 mmol/l) BG 200 + typical

HbA1c < 5.7% 5.7 6.4% HbA1c > 6.5%


insulin secretion), ethnicity, and weight gain are beta-cell dysfunction. The role of beta-cell dys-
major predictors of progression to T2DM. function in predicting progression to T2DM indi-
Longitudinal studies in subjects from a high- cates that interventions that prevent or replenish
risk population (Pima Indians) [41] with baseline the progressive decline in insulin secretion can be
normal glucose tolerance (NGT) indicated that expected to prevent the development of diabetes.
weight gain, insulin resistance, and progressive
loss of insulin secretory response to glucose pre-
dicted the development of T2DM [34]. Weight Predictors of Initial Transition
gain also predicted progression from NGT to IGT to Prediabetes
(5.2 kg vs. 2.6 kg in nonprogressors) and progres-
sion from IGT to T2DM during a 6-year follow-up In contrast to the numerous studies on the transi-
period [34]. The greater weight gain in the pro- tion from prediabetes to T2DM [34, 4042],
gressors was accompanied by ~ 30 % worsening of information on the incidence of prediabetes
insulin resistance and >50 % decline in acute insu- among initially normoglycemic persons is scant.
lin secretory response to intravenous glucose [34]. In a study of 254 Pima Indians with normoglyce-
Weight gain also predicted incident T2DM in mia, 79 subjects (31 %) progressed to prediabetes
African-Americans in the Atherosclerosis Risk in (IGT) during a mean follow-up period of 4 years
Communities study [42]. It follows therefore that [43]. Based on those results, the incidence of pre-
interventions that induce weight loss (e.g., diet, diabetes among Pima Indians can be estimated at
exercise, medications) could prevent progression ~8 %/year [43]. Of the 468 subjects with NGT at
from prediabetes to T2DM. In the prospective enrollment in the Baltimore Longitudinal Study
study of Pima Indians [34], progressive impair- on Aging (BLSA), over half were followed for at
ment of first-phase insulin secretion proved to be a least 10 years [44]. By 10 years, 62 % of the ini-
critical determinant of progression from NGT to tially NGT participants had progressed to predia-
IGT and from IGT to T2DM. Progression from betes, yielding an incidence rate of prediabetes
IGT to diabetes was associated with ~75 % decline 6.2 % in the BLSA cohort (96 % of whom were
in acute insulin secretory response to intravenous European-American) [44]. Dagogo-Jack et al.
glucose [34]. A high concordance rate for impaired followed 343 healthy African-American and
insulin secretion has also been reported among European-American offspring of parents with
elderly identical twins discordant for T2DM (42), T2DM in the Pathobiology of Prediabetes in a
which suggests a genetic basis for pancreatic Biracial Cohort (POP-ABC) study and observed
12 S. Dagogo-Jack

that 100 had developed incident prediabetes (IGT levels [4547]. Obesity is a likely unifying factor
and/or IFG) during a mean follow-up period of that links the various pathophysiological mecha-
~3 years, without evidence of ethnic disparities nisms leading to dysglycemia. Comparison of
[45]. Thus, among black and white subjects with several measures of adiposity indicates higher
parental history of T2DM, the incidence of pre- values in people who progress from normoglyce-
diabetes was ~10 %/year. These data indicate that mia to prediabetes compared with those who
the risk of incident prediabetes among offspring maintain normal glucose metabolism (Table 2.2).
of parents with T2DM in the general United
States population is similar to or higher than the
risk observed among Pima Indians, a group with Rationale for Primary Prevention
the worlds highest rate of T2DM [41]. The POP- of Type 2 Diabetes
ABC data underscore the importance of heredity,
familial, and genetic T2DM. Taken together, There are compelling reasons why the primary
these studies found that normoglycemic individ- prevention of T2DM ought to be an urgent policy
uals develop prediabetes at an annual rate of priority in developing countries. Undoubtedly,
610 %, the higher rate being more likely among the prohibitive costs of managing diabetes and its
those with a strong family history of T2DM. Based complications could easily overwhelm the bud-
on findings in the Pima Indian [34] and the POP- gets of many low- and middle-income countries.
ABC [45] studies, the predictors of incident pre- In many such countries, competing pressures
diabetes include older age, male gender, from infectious diseases and periodic epidemics
overweight/obesity, lower insulin sensitivity, and can easily relegate considerations for diabetes
impaired acute insulin secretory response to glu- care to the bottom of the totem pole of noncom-
cose [45]. Other predictors of incident prediabe- municable diseases. However, the unique demo-
tes included food habits, physical inactivity, graphic (increased diabetes susceptibility at
higher C-reactive protein, and lower adiponectin younger age and female preponderance) and the

Table 2.2 Selected baseline demographic and clinical characteristics of participants who developed prediabetes (pro-
gressors) compared to those who remained free of incident prediabetes during 5 years of follow-up in the POP-ABC
Characteristic Progressors Nonprogressors P-value
Number 111 232
White/black 53/58 97/135 0.3
Female/male 65/46 180/52 0.0003
Age (year) 47 8.9 43.9 10.7 0.0017
Age 1840/4065 23/88 85/147 0.0030
Weight (kg) 90 20 83 22 0.0036
BMI (kg/m2) 31.4 6.9 29.6 7.4 0.0013
Waist (cm) 99 14 92 16 <0.0001
Female 98 12 91 16 0.0006
Male 101 15 96 15 0.14
Total fat mass (kg) 32.0 12.6 29.9 14.0 0.0025
Female 37.2 12.0 32.1 14.4 0.02
Male 24.2 9.1 22.4 9.1 0.0004
Trunk fat mass (kg) 16.6 6.8 14.3 7.3 <0.0001
Female 18.9 6.7 15.1 7.6 0.0064
Male 13.1 5.4 11.5 5.5 0.07
Plusminus values are means SD
BMI body mass index, POP-ABC pathobiology of prediabetes in a biracial cohort (see [47])
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 13

Fig. 2.5 Human

capital investment in
building a
multidisciplinary Endocrinologist/
team-based diabetes Diabetologist
care service. A Ophthalmologist Podiatrist
well-informed and
highly motivated
patient is critical to the
success of diabetes
care Diabetes Patient
Nurse, Educated Dietitian/
Diabetes Motivated Nutritionist
Educator Self-actualizing

Primary Care

geographical (urban vs. rural) distribution of the monitoring supplies, laboratory support, and the
diabetes provide clear targets for preventive professional care team (comprising of physician
intervention. Moreover, there is interplay between specialists, nurses, dietitians, diabetes educators,
diabetes and chronic infectious diseases (e.g., podiatrists, and other experts) is often elusive in
tuberculosis) as well as between successful con- many countries that face challenges in health-
trol of HIV infection and iatrogenic diabetes. care infrastructure (Fig. 2.5). Where some or all
But, the strongest argument for prioritizing dia- of these essential resources for effective diabetes
betes prevention is the premium availability of control are available, their distribution often is
effective tools that have been developed in land- lopsided, and most rural areas are underserved. In
mark studies conducted in different parts of the the absence of comprehensive national health
world that established the feasibility of prevent- coverage, affordability of medications and self-
ing T2DM [4851]. management supplies, even when available, can-
Diabetes mellitus imposes a huge drain on not be guaranteed for most low-income patients.
national health budgets. The annual diabetes- For example, in sub-Saharan Africa, insulin secu-
related health-care costs exceed $240 billion in rity and affordability have been major concerns
the United States [52]. Although precise figures [55, 56]. Lacking cost defrayment by government
are hard to come by, on average, developing or reimbursement by third-party insurers, the
countries spent at least 5 % of their total health limitations imposed by the significant out-of-
expenditures on diabetes in 2010 [53]. The 5 % pocket costs of diabetes care translate to chronic
budgetary allocation is suboptimal and could be suboptimal care, which increases the burden of
much higher if diabetes-related medical costs diabetes-related complications [57].
were to be adjusted to account for the >50 % rate Type 2 diabetes, a global epidemic, now ranks
of undiagnosed diabetes in developing countries among the leading noncommunicable public
[53, 54]. Optimal control of glycemia and related health challenges of the present era [1, 4, 5]. The
comorbidities is difficult and expensive to achieve public health burden imposed by diabetes is
in patients with established diabetes. Availability underscored by the fact that diabetes now is the
of antidiabetes medications, diabetes testing and leading cause of blindness, end-stage renal failure,
14 S. Dagogo-Jack

US (n=1,444) AUS (n=146) BEL (n=173) FRA (n=144) GER (n=256)
GRE (n=176) NOR (n=142) RUS (n=95) SPA (n=182) SWE (n=164)
SWI (n=205) TUR (n=210) UK (n=153)

At Goal,%

50 43.8 41.9
39.7 41.5 40.6 41.8
40 33.7
26.2 26 27.7
30 23.4



Fig. 2.6 Percentage of patients with diabetes achieving optimal control in the United States (HbA1c target <7 %) or
several European countries (HbA1c target <6.5 %) (Source: see [67, 68])

and nontraumatic limb amputations and a major multiple-related risk factors. Consequently, the
contributor to heart disease, stroke, and peripheral long-term complications of diabetes flourish
vascular disease [5760]. These complications unchecked. Ironically, many developing countries
can be prevented or delayed by achieving and lack the infrastructure, technology, and human
maintaining excellent control of glycemia and resources for adequate management of diabetes
comorbid conditions, such as hypertension and complications. Services like dialysis, renal trans-
dyslipidemia [6164]. However, the achievement plantation, laser surgery for retinopathy, interven-
of sustained glycemic control to the level neces- tional cardiology, and rehabilitation services for
sary for prevention of complications often proves amputees are not routinely available. It is, thus,
elusive, even in countries in the developed econo- self-evident that primary prevention of T2DM is
mies [65, 66] (Fig. 2.6). Optimal glycemic control an imperative for developing countries [68, 69].
requires a highly motivated patient, working with
a diabetes care team comprising physicians and
several cadres of clinicians (Fig. 2.5). The care Unique Vulnerabilities
processed involves the use of multiple medica- in Developing Countries
tions; frequent clinic visits; adherence to chal-
lenging lifestyle prescriptions; performance of Data from surveys in developing countries indi-
demanding self-management tasks; paying for cate that diabetes predominantly affects younger
cumulative costs of home blood glucose test age-groups: the majority of people with diabetes
strips; sustained engagement by a team of physi- fall within the age range of 4059 years, as com-
cians, nurses, dietitians, diabetes educators, oph- pared to 60 years or older in developed countries
thalmologists, podiatrists, behaviorists, and other [69]. It has also been predicted that future increases
specialized professionals; and the implementa- in diabetes numbers would affect all age-groups in
tion of sundry other recommendations [13, 67] developing countries, whereas in developed coun-
(Fig. 2.5). Lacking the requisite resources and tries an increase is expected predominantly among
support for excellence in diabetes care, many persons older than 60 years, with a slight decrease
patients in developing countries face the specter in the younger age-groups [69]. This younger age
of chronic suboptimal control of glycemia and predilection means that individuals in developing
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 15

countries in their prime productivity years are the efficacious in decreasing the rate of progression
ones burdened with diabetes and its complica- from prediabetes to T2DM during study periods
tions, with dire consequences on national econo- that ranged from ~3 to 6 years [4851].
mies. Besides the enormous direct medical costs,
the additional lost productivity from absenteeism
and presenteeism inflicts compounding negative Da Qing Study
effects on current and future economic perfor-
mance in the developing world. Furthermore, the Investigators in the Da Qing study (49) screened
preponderance of diabetes in young women of 110,660 men and women from 33 health-care
childbearing age perpetuates a vicious cycle clinics in the city of Da Qing, China, and enrolled
through the effects of intrauterine fetal program- 577 adults (mean age 45 years; mean BMI 26 kg/
ming for increased susceptibility to cardiometa- m2) with IGT. The participants were randomized
bolic disorders in postnatal life [70, 71]. A more by clinic to a control group or to one of three
recent vulnerability is the unexpected association active treatment groups diet only, exercise only,
of successful antiretroviral therapy in HIV or diet plus exercise and were followed every
patients with treatment-emergent metabolic per- 2 weeks during the first 3 months and quarterly
turbations, including diabetes, dyslipidemia, and thereafter for 6 years.
lipodystrophy [72, 73]. The primary end point, development of T2DM
at 6 years, occurred in 67.7 % of subjects in the
control group, 43.8 % in the diet-only group,
Approach to Prevention of Type 2 41.1 % in the exercise-only group, and 46.0 % in
Diabetes the diet-plus-exercise group. Interestingly, rela-
tive decrease in diabetes incidence in the active
Lifestyle Modication treatment groups was similar in lean or over-
weight (BMI >25 kg/m2) subjects. After adjust-
Three landmark studies have demonstrated the ments for baseline differences in BMI and fasting
efficacy of lifestyle intervention in preventing the glucose, the diet, exercise, and diet-plus-exercise
development of T2DM in high-risk individuals interventions produced 31 %, 46 %, and 42 %
[4851]. All studies targeted persons with predia- reductions in diabetes risk, compared with control
betes (principally IGT and high-normal fasting [48]. However, there was apparently no additive
plasma glucose). The lifestyle counseling focused efficacy of combined diet plus exercise versus
on dietary intervention and increased physical either component of the lifestyle intervention.
activity and a weight loss target of approximately
5 to <10 % of initial body weight. The dietary
intervention was aimed at encouraging partici- Finnish Diabetes Prevention Study
pants to adopt healthy eating patterns and to
decrease caloric consumption (by ~500700 kcal/ In the Finnish Diabetes Prevention Study (FDPS)
day) through selective reduction in saturated fat [49], 522 middle-aged subjects (mean age,
calories and limitation of excessive carbohydrate 55 years; mean BMI 31 kg/m2) with IGT were
intake. The physical activity component moti- randomly assigned to either an intervention or
vated high-risk individuals to accrue 150 control group. Each participant in the interven-
240 min of moderate-intensity activity per week tion group received individualized sessions with
(~3060 min daily on 5 days or more each week). a health counselor (approximately every
The target intensity (~55 % VO2 max) of physical 2 months) and was encouraged to aim for ~5 %
activity is equivalent to walking at a brisk pace weight loss through reduction of total and satu-
[4851]. Compared with the control groups (who rated fat intake and increased intake of fiber. The
merely received passive health information), lifestyle participants also were instructed to
these lifestyle modifications proved remarkably increase their physical activity by ~210 min per
16 S. Dagogo-Jack

week. The primary end point was development of After cessation of the active phase of the stud-
T2DM (confirmed by OGTT). The cumulative ies, follow-up assessment showed continued ben-
incidence of diabetes after 4 years was 11 % in efit of lifestyle intervention in decreasing diabetes
the intervention group and 23 % in the control incidence during post-study follow-up periods
group, a significant 58 % reduction in diabetes spanning 1020 years [7577], clearly demon-
incidence. The mean weight loss in the lifestyle strating a legacy effect.
group was ~3.5 kg compared with ~0.8 kg in the
control group. Improvement in insulin and pres-
ervation of insulin secretion also occurred differ- Indian Diabetes Prevention Program
entially in the lifestyle group compared with
control [49]. The Indian Diabetes Prevention Program (IDPP)
randomized South Asians with IGT to four arms:
control (with standard advice, N = 136), lifestyle
Diabetes Prevention Program modification (N = 133), low-dose metformin
(N = 133), and lifestyle modification plus low-
The Diabetes Prevention Program (DPP) enrolled dose metformin (N = 136) [51]. Participants in
3,234 participants with IGT and high-normal lifestyle modification groups received counseling
fasting glucose and assigned them randomly to sessions aimed at promoting healthy eating hab-
intensive lifestyle intervention (ILI), metformin, its (decreased intake of refined carbohydrates and
or placebo treatment [50]. The enrollees included fats and increased intake of dietary fiber) and
representation from all ethnic and racial groups boosting physical activity (at least 30 min daily).
in the US population; persons of non-European After a median follow-up of 30 months, the rela-
ancestry constituted 45 % of the entire cohort. tive reductions in diabetes incidence were 28.5 %
The ILI targets were a minimum of 7 % weight with lifestyle modification, 26.4 % with metfor-
loss/weight maintenance and a minimum of min, and 28.2 % with lifestyle modification and
150 min of physical activity per week. Subjects metformin [71]. Remarkably, these benefits of
in the ILI group received a 16-lesson curriculum lifestyle modification occurred despite the lack of
covering diet, exercise, and behavior modifica- significant weight change in the two groups that
tion delivered by case managers on a one-to-one focused on lifestyle change.
basis during the first 24 weeks after enrollment.
Subsequently, monthly individual sessions and
group sessions with the case managers were pro- Translating Diabetes Prevention
vided, to reinforce the behavioral changes. to Communities in Developing
After an average follow-up period of Countries
2.8 years, the participants randomized to ILI
showed a 58 % reduction in the incidence of dia- Collectively, the extant data show that lifestyle
betes, as compared with placebo [50]. This ben- modification is remarkably and consistently
eficial effect of lifestyle intervention was seen effective in preventing the development of T2DM
in all demographic subgroups defined by age, in high-risk populations in China, India, Europe,
gender, race, or ethnicity. Furthermore, rever- and the United States. To date, no large RCTs of
sion to normal glucose tolerance (NGT) diabetes prevention have been published from
occurred in ~40 % of subjects in the lifestyle Africa, Latin America, Australia, New Zealand,
intervention arm, as compared with ~18 % in the the Middle East, or the Caribbean regions,
control arm [50]. Participants who experienced although a number of pilot projects have been
reversion to NGT (even if transiently) were implemented or are ongoing. Some of these pilot
50 % less likely to develop diabetes during long- projects have reported promising results among
term follow-up, as compared with those who indigenous (Maori) New Zealanders [78]. There
had persistent IGT status [74]. is overwhelming evidence that lifestyle
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 17

modification is a more compelling approach to general public, through demonstrative education

diabetes prevention than medications. However, by convinced professionals; and (4) adoption of
the landmark lifestyle intervention programs that evidenced-based changes in practices and behav-
achieved remarkable success in preventing diabe- iors, triggered by the new knowledge. The pre-
tes were designed as randomized controlled trials ceding construct of translational cascade is
(RCTs) and conducted predominantly at aca- somewhat akin to the sequential steps in
demic medical centers. The protocols of many of Prochaskas transtheoretical model of behavior
the RCTs entailed frequent clinic visits, utilized modification: pre-contemplative, contemplative,
specialized multidisciplinary teams (including preparation, action, and change [83, 84].
physicians, nurses, dietitians, psychologists,
exercise physiologists, and others), and con-
sumed substantial resources and support from Dissemination of Diabetes
institutions and funding agencies. Notably, all Prevention Knowledge
protocol-mandated services and interventions
were provided at no cost to participants in the The preponderance of evidence from randomized
diabetes prevention RCTs, many of whom also clinical trials on the efficacy of lifestyle interven-
received incentives and stipends. Thus, the odds tion in preventing T2DM has established the
were stacked heavily toward success in these rationale for diabetes prevention as self-evident
RCTs, and it is crucial to determine whether the among health professionals. However, the screen-
sterling results obtained in the landmark RCTs ing for prediabetes and institution of prompt life-
could be reproduced in the community, without style counseling have not become routine
all of the inbuilt advantages, specialized profes- practice, even in economically advanced coun-
sionals, and resources. Some community initia- tries. Bridging the hiatus between philosophical
tives are currently underway to determine the acceptance of the idea of diabetes prevention
feasibility of community programs for diabetes among health professionals and the adoption of
prevention [79, 80]. The United States Centers pragmatic steps to detect and act upon high-risk
for Disease Control and Prevention (CDC) has persons requires a methodical approach.
been training and certifying community diabetes The information and knowledge base of health-
prevention personnel under the aegis of the care professional must be steered toward a more
National Diabetes Prevention Program [81]. pragmatic approach to primary prevention of dia-
Programs using trained lay persons to deliver betes through a number of established approaches.
adaptations of the DPP lifestyle intervention to First, curriculum development in primary and
groups (rather than one-on-one sessions) in the high schools should begin to introduce seminal
community have been shown to produce promis- data from the behavioral components of the land-
ing results [79, 80, 82]. mark diabetes prevention studies within a wider
context of instruction in wellness and health pro-
motion. Second, students in schools of medicine,
Keys to Translation nursing, pharmacy, and allied medical fields
should receive exposure to formal instruction in
In general, the keys to the translation of any new the design and key findings of the major diabetes
concept or discovery related to health consist of prevention studies, again, within the wider con-
(1) adequate comprehension, evaluation, and text of wellness and health promotion. Third,
acceptance of the new information within a spe- mastery of the principles and methods of primary
cialized professional community; (2) processing prevention of T2DM ought to be a priority item in
and conversion of the new information into the the core curriculum of residency training pro-
professional knowledge base and dissemination grams for physicians. The latter can be imple-
to a wider circle of professionals and students; mented through lectures and workshops by
(3) diffusion of the conceptual information to the intramural experts and invited speakers. Finally,
18 S. Dagogo-Jack

practicing physicians in all disciplines (including local, national, and international diabetes associ-
internists, family physicians, endocrinologists, ations, can be invaluable during the process of
cardiologists, nephrologists, general and special- program building and facilitation.
ist surgeons), podiatrists, nurse practitioners, and
other primary care providers must have their
knowledge base updated to include the tenets of Strategies for Diabetes Prevention
primary prevention of diabetes. in the Community
Operationally, the increased awareness about
diabetes prevention at the primary care level Five key elements can be distilled from the life-
should lead to increased zeal for screening and style intervention protocols utilized by the DPP
detection of individuals with prediabetes (often and other major diabetes prevention trials. Most
relatives of patients with established T2DM). or all of these strategies can readily be adapted
Once individuals have been diagnosed with pre- for widespread application in the community.
diabetes (Fig. 2.4), appropriate referral for life-
style counseling can follow, as recommended by 1. Selection of persons at risk
the American Diabetes Association [13]. All diabetes prevention trials targeted, screened,
Convinced health-care workers and their fami- and enrolled a defined group of at-risk persons,
lies become conduits for the dissemination and using well-known risk factors for T2DM
diffusion of diabetes prevention ideas in targeted (Table 2.2). The merit of that approach is
segments of the public (hospital communities, underscored by the finding that the participants
established patients and their relatives, schools, randomized to placebo did in fact develop dia-
neighborhoods, social media and outlets, reli- betes at an alarming rate (~12 % per year in
gious forums, etc.). Most human societies have DPP, ~18 % per year in IDPP) [50]. Thus, the
encountered diabetes and are probably precondi- published criteria used for selecting at-risk per-
tioned for receptive attention to diabetes cam- sons for diabetes prevention appear to be of
paigns, having been primed by awareness of the high fidelity and can be adopted for translation
more obvious complications, such as blindness, of diabetes prevention in the general populace.
amputation, and end-stage kidney disease. Such a Specifically, a positive family history of T2DM
preconditioning of societal awareness of diabetes in first-degree relatives, overweight or obesity
and its complications bodes well for the dissemi- (using ethnic-specific BMI cutoffs), and a fast-
nation of ideas regarding the rationale and feasi- ing plasma glucose in the range of 96125 mg/
bility of diabetes prevention. Yet, the fact that dl predict a high yield of eligible individuals for
T2DM can be prevented by modest caloric restric- community diabetes prevention efforts. The
tion and physical activity is yet to enter folklore. appropriate BMI cutoff for identifying over-
That gap in the popular consciousness (despite the weight subjects appears to be >22 kg/m2 for
raging global epidemic of diabetes) presents an Asians compared with >25 kg/m2 for most
enormous opportunity for leadership by health- other ethnicities [85, 86]. The inclusive age
care professionals and civic leaders. To convince range for the published diabetes prevention
and motivate large segments of society for pre- studies was >25 years for DPP and Da Qing
ventive action against diabetes requires coordi- studies, 4065 years for FDPS, and 3055 years
nated efforts at the local, regional, state, national, for IDPP [4851]. None of the published dia-
and international leadership levels. Creative pro- betes prevention studies enrolled individuals
grams anchored by ministries and departments of younger than age 25 years, which is a major
health, information, education, and other agencies limitation, given the increasing prevalence of
would also be important catalysts for public edu- T2DM in children and adolescents [87, 88].
cation, awareness, and action. The expertise and Obesity and physical inactivity are major risk
contributions from philanthropic organizations factors for T2DM in children, as in adults.
and other nongovernmental bodies, especially Other risk factors include female gender,
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 19

ethnicity, family history of T2DM, peripubertal for the prevention of T2DM [4851, 95].
age, and intrauterine exposure to diabetes [89 Walking was the preferred activity for the vast
91]. Because the long-term complications of majority of participants in these studies.
diabetes become established over a >10-year Given the generally low rates of voluntary
period, the epidemic of T2DM in children and physical activity in urbanized communities
adolescents predicts dire consequences for [75, 76], innovative motivational strategies
patients at the prime of their youth [92]. [96, 97] will be required to build physical
Primary prevention of childhood T2DM, there- activity into daily routine, especially in high-
fore, is of utmost public health importance. For risk populations. In clinical trials, the physical
the aforementioned reasons, it is desirable for activity intervention component often required
community diabetes prevention initiatives to exercise physiologists or other skilled staff to
lower the inclusive age for screening at-risk implement. For large-scale community trans-
persons well into the childhood and adolescent lation, these specialists may not be essential,
years. A school-based pilot study sponsored by as lay persons can be trained to deliver physi-
the US National Institutes of Health showed cal activity intervention. However, the infra-
that obesity and diabetes risk factors can be structure required for effective large-scale
decreased by lifestyle intervention in sixth implementation of physical activity interven-
grade pupils [93]. tion well-lit and safe walkways, public
2. Delivery of physical activity intervention parks, biking trail, health and fitness clubs,
The physical activity component of lifestyle etc. may not be readily available in many
intervention in the reported diabetes preven- developing countries, especially at over-
tion studies was of moderate intensity (~55 % crowded urban centers. One innovative
VO2 max) and duration (~30 min daily in the approach to the problem of space for physical
DPP). Out of an abundance of caution, partici- activity is to establish partnerships with
pants in the DPP lifestyle intervention arm schools and houses of religious worship, so
underwent submaximal cardiac stress testing that their large real estate can be utilized for
prior to commencement of the physical activ- diabetes prevention activity after school hours
ity program [94]; however, that was not a rou- and during non-worship days.
tine requirement for physical activity in the The tropical environment (especially, high
majority of diabetes prevention trials [48, 49, temperature, humidity, and torrential rain)
51]. It must be stressed that physical activity constitutes an additional barrier to regular out-
was well-tolerated in the DPP, and no untow- door exercise. Cultural barriers to exercise
ard cardiovascular events or musculoskeletal may also exist in some communities, where
injuries were reported. A routine requirement obesity may be venerated as a sign of well-
for prescreening with cardiac stress would be a being and evidence of freedom from wasting
serious logistical and economic hindrance to diseases (such as malnutrition, HIV/AIDS,
the widespread translation of diabetes preven- and tuberculosis). A less obvious obstacle is a
tion and may not be necessary for the majority cultural mindset among blue collar workers
of free-living individuals who are candidates that associates exercise with the privileged
for diabetes prevention in the community. The elite. These (mis)-perceptions must be con-
DPP exercise goal of 150 min per week was fronted using education and public awareness
similar to that prescribed in the Malmo [95] campaigns. Such grassroots educational cam-
and Da Qing [48] studies but lower than the paigns should emphasize the risks of obesity
210 min per week prescribed in the Finnish and the health benefits of modest increases in
study [49]. The DPP and all other landmark physical activity, thereby shifting the focus
studies have demonstrated the efficacy, tolera- from exercise as indulgent leisure of the bour-
bility, and safety of moderate-intensity physi- geoisie to exercise as an engine of health pro-
cal activity (150210 min per week) as used motion among the proletariat. Civic and local
20 S. Dagogo-Jack

government leadership should collaborate to face-to-face counseling sessions in group for-

create community resources for participatory mat, supplemented by virtual contacts (via
health promotion and wellness. The lowest SMS text messaging, e-mail, or web-based
reported dose of exercise which was effective platforms) could be a less expensive approach
in preventing diabetes was ~150 min/week (or to harnessing the power of contact in diabetes
~30 min/day) [50]; however, significant meta- prevention efforts. Along these lines, web-
bolic benefits can be derived from shorter based lifestyle programs have been reported
bouts of activity (1015 min spread across two to have modest but variable effects on weight
or three periods during the day) [98]. This spe- loss in studies reported from Australia, the
cific point is worth emphasizing when coun- United States, Europe, and Asia [99103];
seling time-pressed subjects. Additionally, however, extrapolation to populations with
whenever appropriate, participatory physical lower literacy and uncertain digital access is
activity and health promotion should be built questionable.
into scheduled cultural festivals and other 4. Delivery of dietary intervention
activities. The latter could include student The participants in the landmark diabetes pre-
teacher and parentchild low- to moderate- vention studies [4851] were asked to reduce
intensity sporting events, parades, pageants, their daily intake of fat calories and total calo-
mini-carnivals, and other civic celebrations. ries. Consumption of meals high in dietary
3. Frequent contacts fiber was also promoted. To help participants
The frequency of contacts between partici- maintain a healthy eating pattern, additional
pants in diabetes prevention programs and life- training regarding reading and understanding
style interventionists was weekly, monthly, food labels was provided. The nutritional
bimonthly, or quarterly, depending on the spe- counseling was delivered in person during the
cific study protocols and phase of study. In the regular visits. As already discussed, there
DPP, the greatest weight loss occurred during appeared to be a relationship between fre-
the initial 24 weeks of intensive individual quency of counseling visits and the magnitude
weekly sessions with lifestyle coaches [50]. of weight loss in the DPP. However, other pro-
When the visit frequency was relaxed to grams that used less frequent visits achieved
monthly sessions, some weight regain was comparable efficacy in diabetes prevention.
noted. Other studies, notably FDPS and Da Thus, following initial run-in and initiation of
Qing, used lower overall contact frequency the dietary intervention, maintenance visits at
(every 2 months to quarterly) and achieved approximately 3-month intervals may be suf-
comparable results to those achieved by the ficient to yield desirable results in a commu-
DPP [48, 49]. Clearly, regular contact at some nity setting.
frequency between interventionists and partic- Pragmatically, implementing dietary inter-
ipants is desirable for successful diabetes pre- vention for diabetes prevention in local com-
vention. Such contacts could create favorable munities in sub-Saharan Africa and other
dynamics and bonding between lifestyle developing regions would be fraught with
coach and participant. Moreover, the repeated challenges. First, of course, is the shortage of
objective recording of weight, waist circum- trained and culturally adapted dietitians and
ference, and other metabolic measures could medical nutritionists. Second, mandatory food
have important motivational effects. However, labeling, the result of legislation passed some
the optimal frequency of physical contacts decades ago in the West, is not yet a routine
necessary for successful diabetes prevention in practice in many developing countries. In fact,
the community is unclear and could well differ nutritional information may be lacking for
by cultural and regional peculiarities. Thus, many staple foods. Thirdly, even where food
such data would need to be determined through labeling exists, low literacy rates mitigate
pilot studies in different communities. Fewer against their comprehension. Furthermore, the
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 21

popularity of informal retail markets (espe- to incorporate some element of self-monitor-

cially for grains, flour, corn, cereal, rice, beans, ing component into a community-based dia-
cassava powder {garri}, and other tuber- betes prevention campaign. The focus of such
derived staples) that do not offer standardized monitoring could be the type and amount
packaging calls for creative approaches to the (serving size) of food intake, daily physical
implementation of caloric control. Thus, effec- activity minutes, and other lifestyle end
tive dietary intervention strategies for diabetes points, as appropriate. In low literacy situa-
prevention in developing countries must antic- tions, the use of domestic surrogates (spouses,
ipate and creatively overcome or circumvent relatives, school age children, neighbors, fam-
these apparent translational obstacles. In that ily friends, etc.) to capture these end points
regard, the value of local pilot and feasibility could be considered. Where feasible, self-
studies cannot be overemphasized. In low lit- monitored data can be transmitted via SMS
eracy regions, a visual (pictorial) approach to text to the Community Diabetes Prevention
dietary modification could be an effective Centers, for tracking and real-time interven-
alternative to text-based formats. The Dietary tional feedback purposes.
Guidelines for Americans, issued by the
United States Department of Agriculture
(USDA), advocates a simple plate method: Community Diabetes Prevention
half of the plate contains fruits and vegetables; Centers
the other half is divided roughly equally into
grains and protein [104]. The plate method The sheer magnitude of the current and projected
[104, 105] is an attractive model that can be escalation of the diabetes epidemic in developing
evaluated as the basis for teaching portion con- countries mandates the broadest possible response
trol and optimal dietary habits in developing at the community level. One practical, efficient
countries. The visual approach uses a graphic approach is the establishment of Community
display of a typical plate, divided into seg- Diabetes Prevention Centers (CDPC) at several
ments that contain desirable food classes in locations on the African continent, the Caribbean
optimal proportions [105]. region, India and South Asia, the Pacific Rim,
5. Self-monitoring South and Central America, Australia and New
Extensive self-monitoring of nutrient intake Zealand, as well as in communities inhabited by
and minutes spent each day in physical activ- high-risk populations in Europe and North
ity (and type of activity) was an integral part America (Fig. 2.7). Such a groundswell of
of the lifestyle intervention in the DPP [50, CDPCs can serve as essential focal points for the
94]. In other studies, the frequent self-moni- dissemination of diabetes prevention practices.
toring of weight and anthropometrics has Operationally, the CDPCs can be administered
been reported to be associated with long- and staffed by the noncommunicable disease
term maintenance of weight loss [106]. The branch of Ministries and Departments of Health or
exact mechanism whereby self-monitoring through partnership with nongovernmental organi-
of eating and exercise behavior predicts good zations. The primary purpose of CDPCs would be
metabolic outcome is unclear. It is plausible to test and implement culturally and regionally
that such a level of involvement in ones appropriate models for the delivery of physical
health could have a beneficial heuristic effect activity and dietary interventions. Referral to
by guiding food choices and lifestyle deci- CDPCs can be based on a risk factor approach that
sions. For most persons in psychic equilib- focuses on genetic (e.g., family history of diabe-
rium, the potential adverse psychological tes) and other risk markers (Table 2.1). Individuals
impact of frequent self-monitoring of nutri- with diagnosed T2DM can be the conduits for
ent intake and physical activity is probably referral of family members to CDPCs for predia-
negligible. For these reasons, it seems prudent betes screening and intervention.
22 S. Dagogo-Jack

Fig. 2.7 Diabetes

prevention strategies for Low-cost, lifestyle intervention models Partnerships: local, regional, national and
delivered by lay persons in the community international Stakeholders
developing countries,
culminating in the
Community support programs: School-based Engagement of the private sector, workplace,
creation of Community physical activity and nutrition education food industry
Diabetes Prevention
Centers (see [68, 69]) Urban designs that facilitate physical activity Laws: Food pricing, labelling, advertising

Training in T2DM prevention (nursing and Mass media: Promote DM knowledge and
medical school curricula) Motivated behavior

Community Diabetes Prevention Centers

Medications for Diabetes placebo [50]. The STOP-NIDDM (Study to

Prevention Prevent Non-Insulin-Dependent Diabetes
Mellitus) employed acarbose as the intervention
Even in the most successful of the randomized drug and demonstrated a 25 % decrease in the rate
controlled trials, the risk reduction for incident of progression to diabetes, compared to placebo
diabetes following lifestyle intervention was [108]. In the XENDOS (XENical in the Prevention
~60 % [4851]. That raises the argument as to of Diabetes in Obese Subjects) study, the pancre-
whether medications could be offered to high- atic lipase inhibitor orlistat (when prescribed in
risk persons who are unable or unwilling to combination with lifestyle modification) resulted
implement lifestyle changes or in whom the latter in a 37 % risk reduction in incident diabetes
have failed to halt glycemic progression. In fact, among subjects with impaired glucose tolerance,
several of the landmark diabetes prevention trials compared with lifestyle intervention alone [109].
also included pharmacological arms, and addi- Thiazolidinedione drugs were tested in the
tional studies have specifically tested medica- DREAM (Diabetes Reduction Assessment with
tions for diabetes prevention. Given the linear Ramipril and Rosiglitazone Medication) [110],
relationship between diabetes risk reduction and ACT NOW (ACTos NOW for the Prevention of
the amount of weight loss (~10 % for every 1 kg Diabetes) [111], and CANOE (CAnadian
weight loss) [50, 107], a medication that enhances Normoglycemia Outcomes Evaluation) [112]
or maintains weight loss would be a rational studies, which showed diabetes risk reduction
adjunct to lifestyle. What follows is a summary rates of >5075 %, compared with placebo. In
of medications that have been studied for the pre- the NAVIGATOR trial, the use of valsartan for
vention of T2DM (Table 2.4). 5 years, along with lifestyle modification, led to
The list of medications that have been tested a relative reduction of 14 % in the incidence of
includes sulfonylureas, metformin, acarbose, orli- diabetes among subjects with IGT, but the effect
stat, rosiglitazone, and pioglitazone (Table 2.3). of nateglinide was not better than placebo [113].
The DPP demonstrated that intervention with The CANOE trial showed that low-dose combi-
metformin decreased the development of diabetes nation therapy with metformin (500 mg twice
in adults with impaired glucose tolerance by 31 % daily) and rosiglitazone (2 mg daily) decreased
[50]. Curiously, the diabetes prevention efficacy incident type 2 diabetes by 66 % compared with
of metformin was observed only in younger, placebo in IGT subjects. Of note, the low doses
obese (BMI >35 kg/m2) individuals; among older were well tolerated, with minimal effect on clini-
or leaner participants, the effect of metformin in cally relevant adverse events of the individual
diabetes risk reduction was no better than that of drugs [112].
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 23

Table 2.3 Diabetes prevention studies using lifestyle modification and medications
Year Study acronym Follow-up Intervention Outcome
1997 Da Qing 6 years Diet + exercise Decrease, 51 %
2001 DPS, Finland 3 years Diet + exercise Decrease, 58 %
2002 DPP 2.8 years Diet+ Ex vs. Met Decrease, 58 %
2002 STOP-NIDDM 3.3 years Acarbose + diet Decrease, 25 %
2004 XENdos 4 years Orlistat + diet Decrease, 37 %
2006 DREAM 3 years Rosiglitazone Decrease, 60 %
2008 ACT NOW 24 years Pioglitazone Decrease, 72 %
2006 IDPP-1 3 years L/S Met Decrease, 2628 %
Met not additive to
2009 IDPP-2 3 years L/S Pio Pio not additive to
2010 Navigator 5 years Nateglinide No effect
Valsartan Decrease, 14 %
2010 CANOE 4 years Rosi+Met Decrease 69 %
Ex exercise, L/S lifestyle, Met metformin, Pio pioglitazone, Rosi troglitazone

Limitations of Medications Table 2.4 Desirable characteristics of an ideal drug for

diabetes prevention
The drugs that have been tested for diabetes pre- Efficacy: should equal or exceed the efficacy of
vention are associated with a range of adverse lifestyle intervention
effects, the need for continuous therapy to main- Mechanism(s): should repair the pathophysiologic
defects that underlie prediabetes
tain their effects, and consequent adherence bar-
Glucoregulation: should normalize glucose
riers. In the studies that tested the effect of metabolism
interruption of metformin, rosiglitazone and pio- Durability: effects should outlast the period of
glitazone therapy for diabetes prevention, no sus- medication exposure
tained effect off medication was observed, which Adiposity: should induce weight loss or be
indicates that those medications did not funda- weight-neutral
mentally improve the underlying pathophysiol- Safety: should have minimal toxicity and require no
safety monitoring
ogy of prediabetes [114116]. Moreover, in the
Tolerability: should be well tolerated, without GI or
Indian Diabetes Prevention Program (IDPP)-1 other adverse effects
and IDPP-2, neither low-dose metformin nor pio- Cost: should cost less than the least expensive drug for
glitazone, when tested in combination with life- diabetes treatment
style modification, achieved additive reduction of Adapted from [28, 69]
diabetes risk, compared to the effects of lifestyle
modification alone [51, 117]. Clearly, the cumu-
lative costs of long-term (probably life-long) and durable medications for diabetes prevention.
therapy with medications (even where generic Indeed, the poor human record of long-term
versions are available and no adverse effects adherence to behavioral recommendations and
occur) can be prohibitive, particularly for devel- the known physiological adaptations that limit
oping countries. weight loss and trigger counter-veiling mecha-
For reasons that have already been argued, the nisms that promote weight regain [118] create a
use of drugs for diabetes prevention cannot be need for such adjunctive pharmacological agents.
recommended as a first-line approach in the gen- The ideal drug for diabetes prevention (Table 2.4)
eral population. The latter conclusion is not to should be nontoxic, well-tolerated, and at least as
eschew societal expectation of safe, effective, efficacious as lifestyle modification [28, 69].
24 S. Dagogo-Jack

Additionally, such a drug should repair or first 6 months (i.e., approximately 1 % per month).
improve the pathophysiological defects that Thus, candidates for diabetes prevention who are
underlie prediabetes, so that a durable effect that unable to meet the lifestyle response target of los-
outlasts the period of medication can be expected. ing ~1 %/month of body weight during the initial
The latter attribute would permit withdrawal of 36 months of behavioral intervention may be
the medication after a defined period of interven- considered for adjunctive metformin therapy.
tion, without the risk of prediabetes relapse. The guidelines on the management of predia-
Finally, the cost of such a drug must not be pro- betes, issued by the Indian Health Services (IHS)
hibitive, bearing in mind the large number of and the Australian Diabetes Society/Australian
people with prediabetes (86 million in the United Diabetes Educators Association, emphasize life-
States and more than 400 million worldwide). style intervention prior to consideration of medi-
cations. The Australian guidelines recommend
trying lifestyle intervention for a minimum of
Current Guidelines for Use 6 months before considering drugs for diabetes
of Medication for Diabetes prevention [120]. The IHS guidelines recom-
Prevention mend addition of either metformin or piogli-
tazone if initial lifestyle intervention fails to
The ADA consensus statement [119] recom- improve dysglycemia in people with prediabetes
mends lifestyle modification with a weight loss [121]. The IHS guidelines state that the decision
goal of 510 % along with moderate physical to use medication for diabetes prevention must be
activity of about 30 min daily for patients with made on an individual basis and with the patients
IFG or IGT. Although no drug has been approved full understanding [121].
by the Food and Drug Administration for diabe- Prediabetes is diagnosed based on the presence
tes prevention, the ADA has suggested that treat- of either IFG alone, IGT alone (determined during
ment with metformin be considered as an adjunct oral glucose tolerance test), or both IFG and
to diet and exercise for the prevention of type 2 IGT. Individuals who have both IFG and IGT (so-
diabetes in selected high-risk persons [13, 119]. called double prediabetes) show more severe insu-
Based on the subgroup analysis of the efficacy of lin resistance and impairment of beta-cell function
metformin in the DPP, metformin was most compared to persons with a single prediabetes
effective in preventing diabetes in high-risk, very marker [119, 122, 123]. Nearly all persons with
obese (BMI >35 kg/m2) prediabetic subjects double prediabetes (~96 %) would qualify for met-
younger than 60 years of age [50]. Additional formin therapy, based on the ADA consensus cri-
selection criteria when considering metformin teria, whereas only ~30 % of persons with isolated
use in prediabetic subjects include a family his- IFG would be eligible for metformin treatment
tory of diabetes in first-degree relatives, prior using the same criteria [119, 124]. This means that
gestational diabetes mellitus, hypertriglyceride- the use of oral glucose tolerance testing can refine
mia, subnormal HDL cholesterol levels, hyper- and sharply tailor the selection of individuals at
tension, and HbA1c 5.76.4 % [13, 119]. greatest risk (i.e., those with IFG + IGT) for met-
Even in persons who harbor all or most of these formin adjunctive treatment [119, 124]. This
risk factors, active lifestyle modification is the pre- maximalist risk and minimalist drug interven-
ferred initial intervention; metformin can then be tion approach has much to commend it, as it would
considered for individuals who fail to make sig- spare individuals and governments the huge
nificant progress. Currently, there are no clear expense of covering the costs of medications and
guidelines for determining the optimal timing of related expenses for a much wider range of indi-
metformin therapy, but failure of lifestyle inter- viduals from lower risk pools. Indeed, the ADA
vention can be determined fairly empirically. The consensus statement stipulates that the presence of
DPP participants assigned to lifestyle intervention both IFG and IGT must be documented if metfor-
lost ~7 % of their baseline body weight during the min is to be used for diabetes prevention [119].
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 25

Costs of Preventing T2DM agencies) would need to be explored and mobi-

lized for successful implementation of diabetes
Analysis of costs associated with the DPP indi- prevention initiatives in most low- and middle-
cates that the overall expense to society of pre- income countries.
venting diabetes was cost-effective [125128].
For over 3 years, the direct medical costs of the
DPP interventions were US $79 per participant Conclusion
in the placebo group, US $2,542 in the metfor- Nearly 90 million adults in the United States
min group, and US $2,780 in the lifestyle group and more than 400 million people around the
[125, 126]. Further longer-term analysis indi- world have prediabetes. Perhaps, one of the
cated that lifestyle modification remained cost- most significant public health advances of the
effective, and metformin was potentially cost present era is the demonstration that progres-
saving when used for preventing diabetes [128]. sion from prediabetes to type 2 diabetes can
Cost analysis in the IDPP showed that the be interrupted by effective interventions. In
T2DM prevention strategies in India, especially addition to the compelling arguments for pri-
lifestyle intervention, were cost-effective [129]. oritizing diabetes prevention, there is an even
The total cost of identifying one person with more urgent need for action in low-income
IGT was US $117. The direct medical costs countries, where a narrow window of opportu-
were US $61 per person in the control group and nity currently exists. In those countries, diabe-
US $225 in the lifestyle intervention group tes is currently not featured among the top ten
[129]. The cost of preventing one case of T2DM causes of death; however, as countries emerge
through lifestyle modification in India was US from poverty, the role of diabetes as a leading
$1,052 [129]. The DPP, IDPP, and other trials cause of death becomes firmly established
were by the nature of their design quite (Fig. 2.8). This nefarious association between
resource-intensive. economic progress and death from diabetes is
Clearly, the adaptation of diabetes prevention not inevitable; economic transition from pov-
programs to developing countries would require erty to wealth can and must be uncoupled
extensive cost-containment and cost-shifting from increased morbidity and mortality from
strategies. Reduction in contact frequency, diabetes [131]. Among the different interven-
optimization of the size of the intervention per- tions of proven efficacy in preventing diabe-
sonnel, and the use of group (rather than one-on- tes, lifestyle modification is the most appealing
one) counseling format should decrease costs because of its high efficacy, nontoxicity, and
significantly. Furthermore, the use of trained generalizable effects in persons from various
lay (nonmedical) workers to implement life- ethnic backgrounds [4851]. Most of the med-
style modification protocols in group sessions ications that have been tested for diabetes pre-
at the Community Diabetes Prevention Centers vention fall short of the efficacy achieved by
and prorated cost sharing could be additional lifestyle intervention. Although the efficacy of
approaches to the implementation of afford- thiazolidinedione drugs on diabetes preven-
able diabetes prevention programs in develop- tion when administered to persons with pre-
ing countries [79, 130]. Even with a markedly diabetes can match or somewhat exceed that
scaled down cost schedule, the low gross domes- of lifestyle intervention, the drug effect is not
tic product (~$500 per capita) in many develop- sustained and quickly dissipates following
ing countries suggests that these nations cannot cessation of therapy [114, 115]. In contrast
afford to underwrite the costs of even a limited such a rapid washout effect is not seen with
national diabetes prevention program. Given the lifestyle intervention, where sustained bene-
immense future benefits to society from preven- fits of diabetes prevention have been reported
tion of T2DM, novel funding sources (private, up to 20 years following cessation of formal
industry, international, philanthropic, and other intervention [7577].
26 S. Dagogo-Jack

Lower respira- Ischaemic

91 95
tory infections heart disease
HIV/AIDS 65 Stroke 78
Diarrhoeal Lower respira-
53 53
diseases tory infections
Stroke 52 COPD 52
e e s
m Diarrhoeal dl ie
heart disease
i n s diseases
id ntr
m u
w- rie Preterm r - o
Malaria 35
Lo unt birth complic...
we e c
33 co HIV/AIDS 23
Lo com
birth complic...
Tuberculosis 31 22
29 Tuberculosis 21
asphyxia and...
Protein Cirrhosis of
27 19
energy maln... the liver
0 20 40 60 80 100 0 20 40 60 80 100
Deaths per 100,000 population Deaths per 100,000 population
158 Stroke 126
heart disease
Stroke 95 107
heart disease
49 COPD 50
bronchus, lu...
Alzheimer Trachea
42 31
Disease and ... e bronchus, lu...
e ies
m dl
COPD 31 co Diabetes
id ntr
- in s mellitus
m u
Lower respira-
gh rie Lower respira- r- o
tory infections
Hi unt tory infections
pe e c
27 co Road injury 21
Up com
rectum cance...
Diabetes Hypertensive
20 20
mellitus heart disease
Hypertensive Liver
20 18
heart disease cancer
Breast Stomach
16 17
cancer cancer
0 32 64 96 128 160 0 26 52 78 104 130
Deaths per 100,000 population Deaths per 100,000 population

Fig. 2.8 Top ten causes of death in countries of the world of death. COPD chronic obstructive pulmonary disease
grouped by income status. Diabetes does not appear (Source: World Health Organization. Available from
among the top causes of death in low-income countries,
but as countries experience upward economic transition, index1.html. Accessed 20 May 2015)
diabetes becomes firmly established among the top causes

The effective translation of diabetes pre- Diabetes Prevention Centers, to serve as a

vention across global communities requires hub for harmonizing the planning, delivery,
engagement and coordination at multiple and evaluation of lifestyle intervention activi-
levels within the health-care establishment, ties, would be an efficient approach to global
civic society, and government agencies. diabetes prevention. Appropriate legislation
Environmental and policy changes are needed that promotes food labeling, rewards healthful
to stimulate broad societal participation in well- behavior, and redistributes revenue from sin
ness and diabetes prevention activities. Pilot tax on tobacco and alcohol to fund community
and feasibility projects are needed to custom- diabetes prevention initiatives would be well
ize generally proven methodologies to local directed. Clearly, great opportunities exist for
specificities and realities in developing coun- innovative partnerships among governmental,
tries and underserved communities. Ecological civic, philanthropic, and other nongovernmen-
improvements that result in improved access tal organizations toward a common purpose
to safe walking trails, well-lit and well-kept of stemming the global diabetes epidemic by
public parks, subsidized or affordable neigh- focusing on primary prevention.
borhood fitness centers, and other appurte-
nances for health promotion would all augur Acknowledgments SD-J is supported, in part, by grants
well for the practice of community diabetes (R01 DK067269, DK62203, DK48411) from the National
prevention. The establishment of Community Institutes of Health.
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 27

Disclosures The author has no conflicts of inter- 15. Dagogo-Jack S, Santiago JV. Pathophysiology of
est with regard to the content of this chapter. type 2 diabetes and modes of action of therapeutic
interventions. Arch Intern Med. 1997;157:
16. Defronzo RA. Banting lecture. From the triumvirate to
References the ominous octet: a new paradigm for the treatment of
type 2 diabetes mellitus. Diabetes. 2009;58:77395.
1. International Diabetes Federation. IDF diabetes atlas 17. DeFronzo RA. Insulin resistance, lipotoxicity, type 2
2015. 7th ed. Brussels: International Diabetes diabetes and atherosclerosis: the missing links. The
Federation; 2015. Claude Bernard Lecture 2009. Diabetologia. 2010;
2. Geiss LS, Wang J, Cheng YJ, Thompson TJ, Barker L, 53:127087.
Li Y, et al. Prevalence and incidence trends for diag- 18. Rasmussen BB, Holmbck UC, Volpi E, Morio-
nosed diabetes among adults aged 20 to 79 years, Liondore B, Paddon-Jones D, Wolfe RR. Malonyl
United States, 19802012. JAMA. 2014;312:121826. coenzyme A and the regulation of functional carni-
3. Gregg EW, Li Y, Wang J, Burrows NR, Ali MK, tine palmitoyltransferase-1 activity and fat oxidation
Rolka D, et al. Changes in diabetes-related compli- in human skeletal muscle. J Clin Invest. 2002;
cations in the United States, 19902010. N Engl 110:168793.
J Med. 2014;370:151423. 19. Kitabchi AE, Temprosa M, Knowler WC, Kahn SE,
4. World Health Organization. Global status report on Fowler SE, Haffner SM, et al. Role of insulin secre-
noncommunicable diseases 2014. Geneva: WHO tion and sensitivity in the evolution of type 2 diabe-
Press; 2014. tes in the diabetes prevention program: effects of
5. Shaw JE, Sicree RA, Zimmet PZ. Global estimates lifestyle intervention and metformin. Diabetes.
of the prevalence of diabetes for 2010 and 2030. 2005;54:240414.
Diabetes Res Clin Pract. 2010;87:414. 20. Larson-Meyer DE, Heilbronn LK, Redman LM,
6. Hu F. Globalization of diabetes: the role of diet, life- Newcomer BR, Frisard MI, Anton S, et al. Effect of
style, and genes. Diabetes Care. 2011;34:124957. calorie restriction with or without exercise on insulin
7. Egede LE, Dagogo-Jack S. Epidemiology of type 2 sensitivity, -cell function, fat cell size, and ectopic
diabetes: focus on ethnic minorities. Med Clin N lipid in overweight subjects. Diabetes Care. 2006;
Am. 2005;89:94975. 29:133744.
8. Tanizawa Y, Riggs AC, Dagogo-Jack S, Vaxillaire 21. Kelley DE. Effects of weight loss on glucose homeo-
M, Froguel P, Liu L, et al. Isolation of the human stasis in NIDDM. Diabetes Rev. 1995;3:36677.
LIM/homeodomain gene islet-1 and identification of 22. Schneider SH, Morgado A. Effects of fitness and
a simple sequence repeat polymorphism. Diabetes. physical training on carbohydrate metabolism and
2004;43:93541. associated cardiovascular risk factors in patients
9. Ehm MG, Karnoub MC, Sakul H, Gottschalk K, with diabetes. Diabetes Rev. 1995;3:378407.
Holt DC, Weber JL, American Diabetes Association 23. Rogers MA, Yamamoto C, King DS, Harberg JM,
GENNID Study Group. Genetics of NIDDM, et al. Ensani AA, Holloszy JO. Improvement in glucose
Genome wide search for type 2 diabetes susceptibil- tolerance after one week of exercise in patients with
ity genes in four American populations. Am J Hum mild NIDDM. Diabetes Care. 1988;11:6138.
Genet. 2000;66:187181. 24. Eriksson KF, Lindgarde F. Prevention of type 2 dia-
10. McCarthy M, Zeggini E. Genome-wide association betes mellitus by diet and physical exercise.
studies in type 2 diabetes. Curr Diab Rep. 2009;9: Diabetologia. 1991;34:8918.
16471. 25. Mikines KJ, Sonne B, Farrell PA, Tronier B, Galbo
11. Hivert MF, Jablonski KA, Perreault L, Saxena R, H. Effect of physical exercise on sensitivity and
McAteer JB, Franks PW, et al. Updated genetic responsiveness to insulin in humans. Am J Physiol.
score based on 34 confirmed type 2 diabetes Loci is 1988;254:E24859.
associated with diabetes incidence and regression to 26. Goodyear LJ, Kahn BB. Exercise, glucose transport,
normoglycemia in the diabetes prevention program. and insulin sensitivity. Annu Rev Med. 1998;
Diabetes. 2011;60:13408. 49:23561.
12. Dagogo-Jack S. Predicting diabetes: our relentless 27. Steven S, Hollingsworth KG, Al-Mrabeh A, Avery
quest for genomic nuggets. Diabetes Care. 2012; L, Aribisala B, Caslake M, et al. Very-low-calorie
35:1935. diet and 6 months of weight stability in type 2 diabe-
13. American Diabetes Association. Standards of medi- tes: pathophysiologic changes in responders and
cal care in diabetes2016. Diabetes Care. 2016; nonresponders. Diabetes Care. 2016. pii: dc151942.
39:S1112. [Epub ahead of print].
14. Fujimoto WY, Leonetti DL, Kinyoun JL, Newell- 28. Dagogo-Jack S, Edeoga C. Understanding and iden-
Morris L, Shuman WP, Stolov WC, et al. Prevalence tifying pre-diabetes can we halt the diabetes epi-
of diabetes mellitus and impaired glucose tolerance demic? Eur Endocrinol. 2008;4:168.
among second-generation Japanese-American men. 29. Pour OR, Dagogo-Jack S. Prediabetes as a therapeu-
Diabetes. 1987;36:7219. tic target. Clin Chem. 2011;57:21520.
28 S. Dagogo-Jack

30. Maschirow L, Khalaf K, Al-Aubaidy HA, Jelinek 46. Boucher AB, Adesanya EAO, Owei I, Gilles AK,
HF. Inflammation, coagulation, endothelial dysfunc- Ebenibo S, Wan J, Edeoga C, Dagogo-Jack
tion and oxidative stress in prediabetes biomark- S. Dietary habits and leisure-time physical activity
ers as a possible tool for early disease detection for in relation to adiposity, dyslipidemia, and incident
rural screening. Clin Biochem. 2015;48:5815. dysglycemia in the Pathobiology of Prediabetes in
31. Nyenwe EA, Dagogo-Jack S. Metabolic syndrome, A Biracial Cohort Study. Metabolism. 2015;64:
prediabetes and the science of primary prevention. 10607.
Minerva Endocrinol. 2011;36:12945. 47. Jiang Y, Owei I, Wan J, Ebenibo S, Dagogo-Jack
32. Tabk A, Herder C, Rathmann W, Brunner E, S. Adiponectin levels predict prediabetes risk: the
Kivimki M. Prediabetes: a high-risk state for diabe- pathobiology of prediabetes in a biracial cohort
tes development. Lancet. 2012;379:227990. (POP-ABC) study. BMJ Open Diabetes Res Care.
33. Weir G, Bonner-Weir S. Five stages of evolving pro- 2016;4:e000194. doi:10.1136/bmjdrc-2016.
gression to diabetes. Diabetes. 2004;53:S1621. 48. Pan XR, Li GW, Hu YH, Wang JX, Yang WY, An
34. Weyer C, Bogardus C, Mott DM, Pratley R. The ZX, et al. Effects of diet and exercise in preventing
natural history of insulin secretory dysfunction and NIDDM in people with impaired glucose tolerance:
insulin resistance in the pathogenesis of type 2 dia- the Da Qing IGT and Diabetes Study. Diabetes Care.
betes mellitus. J Clin Invest. 1999;104:78794. 1997;20:53744.
35. Toft-Nielsen MB, Damholt MB, Madsbad S, Hilsted 49. Tuomilehto J, Lindstrom J, Eriksson J, Valle T,
LM, Hughes TE, Michelsen BK, et al. Determinants Hamalainen H. Prevention of type 2 diabetes melli-
of the impaired secretion of glucagon-like peptide- tus by changes in lifestyle among subjects with
1 in type 2 diabetic patients. J Clin Endocrinol impaired glucose tolerance. N Engl J Med. 2001;344:
Metab. 2001;86:371723. 134350.
36. Papaetis G. Incretin-based therapies in prediabetes: 50. DPP Research Group. Reduction in the incidence of
current evidence and future perspectives. World type 2 diabetes with lifestyle intervention or metfor-
J Diabetes. 2014;5:81734. min. N Engl J Med. 2002;346:393403.
37. Lopategi A, Lpez-Vicario C, Alcaraz-Quiles J, 51. Ramachandran A, Snehalatha C, Mary S, Mukesh B,
Garca-Alonso V, Titos E, Clria J, et al. Role of bio- Bhaskar AD, Vijay V, Indian Diabetes Prevention
active lipid mediators in obese adipose tissue inflam- Programme (IDPP). The Indian Diabetes Prevention
mation and endocrine dysfunction. Mol Cell Programme shows that lifestyle modification and
Endocrinol. 2016;419:4459. metformin prevent type 2 diabetes in Asian Indian
38. Cerasi E, Luft R. The prediabetic state, its nature and subjects with impaired glucose tolerance (IDPP-1).
consequences a look toward the future. Diabetes. Diabetologia. 2006;49:28997.
1972;21 Suppl 2:68594. 52. American Diabetes Association. Economic costs of
39. Centers for Disease Control and Prevention. National diabetes in the U.S. in 2012. Diabetes Care. 2013;36:
diabetes statistics report: estimates of diabetes and 103346.
its burden in the United States, 2014. Atlanta: US 53. Zhang P, Zhang X, Brown J, Vistisen D, Sicree R,
Department of Health and Human Services; 2014. Shaw J, Nichols G. Global healthcare expenditure on
40. Edelstein SL, Knowler WC, Bain RP, et al. Predictors diabetes for 2010 and 2030. Diabetes Res Clin Pract.
of progression from impaired glucose tolerance to 2010;87:293301.
NIDDM: an analysis of six prospective studies. 54. Barcel A, Aedo C, Rajpathak S, Robles S. The cost
Diabetes. 1997;46:70110. of diabetes in Latin America and the Caribbean. Bull
41. Knowler WC, Bennett PH, Hamman RF, Miller World Health Organ. 2003;81:1927.
M. Diabetes incidence and prevalence in Pima 55. Deeb LC, Tan MH, Alberti KGMM. Insulin availabil-
Indians: a 19-fold greater incidence than in Rochester, ity among International Diabetes Federation member
Minnesota. Am J Epidemiol. 1978;108:497505. associations. Diabetes Care. 1994;17:2203.
42. Brancati FL, Kao WH, Folsom AR, Watson RL, 56. Dagogo-Jack S. DCCT results and diabetes care in
Szklo M. Incident type 2 diabetes mellitus in African developing countries. Diabetes Care. 1995;18:4167.
American and white adults: the Atherosclerosis Risk 57. Beran D, Yudkin JS. Diabetes care in sub-Saharan
in Communities Study. JAMA. 2000;283:22539. Africa. Lancet. 2006;368:168995.
43. Weyer C, Tataranni PA, Bogardus C, Pratley 58. Dagogo-Jack S. Pattern of foot ulcer in diabetic
R. Diabetes Care. 2000;24:8994. Nigerians. Pract Diabetes Dig. 1991;2:758.
44. Meigs JB, Muller DC, Nathan DM, Blake DR, 59. Dagogo-Jack S. Diabetic in-patient mortality in
Andres R. Diabetes. 2003;52:147584. Nigeria. Pract Diabetes Dig. 1991;2:1179.
45. Dagogo-Jack S, Edeoga C, Ebenibo S, Nyenwe E, 60. Dagogo-Jack S. Primary prevention of cardiovascu-
Wan J. Lack of racial disparity in incident prediabe- lar disease in diabetic patients. Cardiol Q. 2006;12:
tes and glycemic progression among black and white 205.
offspring of parents with type 2 diabetes: the patho- 61. Nathan DM, Bayless M, Cleary P, Genuth S,
biology of prediabetes in a biracial cohort (POP- Gubitosi-Klug R, Lachin JM, DCCT/EDIC Research
ABC) study. J Clin Endocrinol Metab. 2014;99: Group, et al. Diabetes control and complications
E107887. trial/epidemiology of diabetes interventions and
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 29

complications study at 30 years: advances and con- follow-up of the Finnish Diabetes Prevention Study.
tributions. Diabetes. 2013;62:397686. Lancet. 2006;368:16739.
62. UK Prospective Diabetes Study (UKPDS) Group. 77. Li G, Zhang P, Wang J, An Y, Gong Q, Gregg EW,
Intensive blood-glucose control with sulphonylureas et al. Cardiovascular mortality, all-cause mortality,
or insulin compared with conventional treatment and and diabetes incidence after lifestyle intervention for
risk of complications in patients with type 2 diabetes people with impaired glucose tolerance in the Da
(UKPDS 33). Lancet. 1998;352:83753. Qing Diabetes Prevention Study: a 23-year follow-up
63. Hayward RA, Reaven PD, Wiitala WL, Bahn GD, study. Lancet Diabetes Endocrinol. 2014;2:47480.
Reda DJ, Duckworth, VADT Investigators, et al. 78. Simmons D, Rush E, Crook N. Development and
Follow-up of glycemic control and cardiovascular piloting of a community health worker-based inter-
outcomes in type 2 diabetes. N Engl J Med. 2015; vention for the prevention of diabetes among New
372:2197206. Zealand Maori in Te Wai o Rona: diabetes preven-
64. Gaede P, Lund-Andersen H, Parving HH, Pedersen tion strategy. Public Health Nutr. 2008;11:131825.
O. Effect of a multifactorial intervention on mortality 79. Ackermann RT, Marrero DG. Adapting the Diabetes
in type 2 diabetes. N Engl J Med. 2008;358:58091. Prevention Program lifestyle intervention for deliv-
65. Banegas JR, Lpez-Garca E, Dallongeville J, ery in the community: the YMCA model. Diabetes
Guallar E, Halcox JP, Borghi C, et al. Achievement Educ. 2007;33(69):745, 778.
of treatment goals for primary prevention of cardio- 80. Albright AL, Gregg EW. Preventing type 2 diabetes
vascular disease in clinical practice across Europe: in communities across the U.S.: the National
the EURIKA study. Eur Heart J. 2011;32:214353. Diabetes Prevention Program. Am J Prev Med.
66. Ali MK, Bullard KM, Saaddine JB, Cowie CC, 2013;44(4 Suppl 4):S34651.
Imperatore G, Gregg EW. Achievement of goals in 81. Albright A. The National Diabetes Prevention
U.S. diabetes care, 19992010. New Engl J Med. Program: from research to reality. Diabetes Care
2013;368:161324. Educ Newsl. 2012;33(4):47.
67. Dagogo-Jack S. Preventing diabetes-related morbid- 82. Hays LM, Finch EA, Saha C, Marrero DG,
ity and mortality in the primary care setting. J Natl Ackermann RT. Effect of self-efficacy on weight
Med Assoc. 2002;94:54960. loss: a psychosocial analysis of a community-based
68. Dagogo-Jack S. Primary prevention of type 2 diabe- adaptation of the diabetes prevention program life-
tes in developing countries. J Natl Med Assoc. style intervention. Diabetes Spectr. 2014;27:2705.
2006;98:4159. 83. Velicer WF, Brick LA, Fava JL, Prochaska
69. Echouffo-Tcheugui JB, Dagogo-Jack S. Preventing JO. Testing 40 predictions from the transtheoretical
diabetes mellitus in developing countries. Nat Rev model again. With confidence. Multivar Behav Res.
Endocrinol. 2012;8:55762. 2013;48:22040.
70. Dabelea D, Knowler WC, Pettitt DJ. Effect of dia- 84. Norcross JC, Krebs PM, Prochaska JO. Stages of
betes in pregnancy on offspring: follow-up research change. J Clin Psychol. 2011;67:14354.
in the Pima Indians. J Matern Fetal Med. 2000; 85. Pan WH, Flegal KM, Chang HY, Yeh WT, Yeh CJ,
9:838. Lee WC. Body mass index and obesity-related meta-
71. Francis-Emmanuel PM, Thompson DS, Barnett AT, bolic disorders in Taiwanese and US whites and
Osmond C, Byrne CD, Hanson MA, et al. Glucose blacks: implications for definitions of overweight
metabolism in adult survivors of severe acute malnu- and obesity for Asians. Am J Clin Nutr. 2004;79:
trition. J Clin Endocrinol Metab. 2014;99:223340. 319.
72. Dagogo-Jack S. HIV therapy and diabetes risk 86. Yang W, Lu J, Weng J, Jia W, Ji L, Xiao J. Prevalence
(Editorial). Diabetes Care. 2008;31:12678. of diabetes among men and women in China. N Engl
73. Yarasheski K, Tebas P, Sigmund CM, Dagogo-Jack S, J Med. 2010;362:1090101.
Bohrer A, Turk J, Cryer PE, Powderly WG. Insulin 87. Pohl JH, Greer JA, Hasan KS. Type 2 diabetes mel-
resistance in HIV-protease inhibitor-associated diabe- litus in children. Endocr Pract. 1998;4:4136.
tes. J Acquir Immune Defic Syndr. 1999;21:20916. 88. Kaufman FR. Type 2 diabetes mellitus in children
74. Perreault L, Pan Q, Mather KJ, Watson KE, Hamman and youth: a new epidemic. J Pediatr Endocrinol
RF, Kahn SE, Diabetes Prevention Program Research Metab. 2002;15 Suppl 2:73744.
Group. Effect of regression from prediabetes to normal 89. Alberti G, Zimmet P, Shaw J, Bloomgarden Z,
glucose regulation on long-term reduction in diabetes Kaufman F, Silink M, Consensus Workshop Group.
risk: results from the Diabetes Prevention Program Type 2 diabetes in the young: the evolving epidemic:
Outcomes Study. Lancet. 2012;379:224351. the international diabetes federation consensus
75. Diabetes Prevention Program Research Group. workshop. Diabetes Care. 2004;27:1798811.
10-year follow-up of diabetes incidence and weight 90. Wiegand S, Maikowski U, Blankenstein O,
loss in the Diabetes Prevention Program Outcomes Biebermann H, Tarnow P, Gruters A. Type 2 diabetes
Study. Lancet. 2009;374:167786. and impaired glucose tolerance in European children
76. Lindstrm J, Ilanne-Parikka P, Peltonen M, Aunola and adolescents with obesity - a problem that is no
S, Eriksson JG, Hemi K. Sustained reduction in the longer restricted to minority groups. Eur
incidence of type 2 diabetes by lifestyle intervention: J Endocrinol. 2004;151:19920.
30 S. Dagogo-Jack

91. Caprio S. Development of type 2 diabetes mellitus in a visual method of teaching meal planning. DAIS
the obese adolescent: a growing challenge. Endocr Project Group. Diabetes Atherosclerosis Intervention
Pract. 2012;18:7915. Study. J Am Diet Assoc. 1998;98:11558.
92. TODAY Study Group. Rapid rise in hypertension and 106. Wing RR, Phelan S. Long-term weight loss mainte-
nephropathy in youth with type 2 diabetes: the TODAY nance. Am J Clin Nutr. 2005;82(Suppl):222S5.
clinical trial. Diabetes Care. 2013;36:173541. 107. Hamman RF, Wing RR, Edelstein SL, Lachin JM,
93. HEALTHY Study Group. A school-based interven- Bray GA, Delahanty L, Hoskin M, Kriska AM,
tion for diabetes risk reduction. N Engl J Med. Mayer-Davis EJ, Pi-Sunyer X, Regensteiner J,
2010;363:4435. Venditti B, Wylie-Rosett J, Diabetes Prevention
94. The Diabetes Prevention Program Research Group. Program Research Group. Effect of weight loss with
The Diabetes Prevention Program (DPP): descrip- lifestyle intervention on risk of diabetes. Diabetes
tion of lifestyle intervention. Diabetes Care. 2002;25: Care. 2006;29:21027.
216571. 108. Chiasson JL, Josse RG, Gomis R, Hanefeld M,
95. Eriksson KF, Lindgarde F. Prevention of type 2 Karasik A, Laakso M, STOP-NIDDM Trial Research
(noninsulin-dependent) diabetes mellitus by diet and Group. Acarbose for prevention of type 2 diabetes
physical exercise: the 6-year Malmo feasibility mellitus: the STOP-NIDDM randomised trial.
study. Diabetologia. 1991;4:8918. Lancet. 2002;359:20727.
96. Pate RR, Pratt M, Blair SN, Haskell WL, Macera 109. Orgerson JS, Hauptman J, Boldrin MN, Sjstrm
CA, Bouchard C, et al. Physical activity and public L. XENical in the prevention of Diabetes in Obese
health: recommendation from the Centers for Disease Subjects (XENDOS) study: a randomized study of
Control and Prevention and the American College of orlistat as an adjunct to lifestyle changes for the pre-
Sports Medicine. JAMA. 1995;273:4027. vention of type 2 diabetes in obese patients. Diabetes
97. Kimm SY, Glynn NW, Kriska AM, Barton BA, Care. 2004;27:15561.
Kronsberg SS, Daniels SR, et al. Decline in physical 110. DREAM (Diabetes REduction Assessment with
activity in black and white girls during adolescence. ramipril and rosiglitazone Medication) Trial
N Engl J Med. 2002;347:70915. Investigators. Effect of rosiglitazone on the fre-
98. Hood MS, Little JP, Tarnopolsky MA, Myslik F, quency of diabetes in patients with impaired glucose
Gibala MJ. Low-volume interval training improves tolerance or impaired fasting glucose: a randomised
muscle oxidative capacity in sedentary adults. Med controlled trial. Lancet. 2006;368:1096105.
Sci Sports Exerc. 2011;43:184956. 111. DeFronzo RA, Tripathy D, Schwenke DC, Banerji
99. Tate DF, Wing RR, Winett RA. Using Internet tech- M, Bray GA, Buchanan TA, ACT NOW Study, et al.
nology to deliver a behavioral weight loss program. Pioglitazone for diabetes prevention in impaired glu-
JAMA. 2001;285:11727. cose tolerance. N Engl J Med. 2011;364:110415
100. Neve M, Morgan PJ, Jones PR, Collins (Erratum in: N Engl J Med. 2011;365:189).
CE. Effectiveness of web-based interventions in 112. Zinman B, Harris SB, Neuman J, Gerstein HC,
achieving weight loss and weight loss maintenance Retnakaran RR, Raboud J, et al. Low-dose combina-
in overweight and obese adults: a systematic review tion therapy with rosiglitazone and metformin to
with meta-analysis. Obes Rev. 2010;11:30621. prevent type 2 diabetes mellitus (CANOE trial): a
101. Kodama S, Saito K, Tanaka S, Horikawa C, Fujiwara double-blind randomised controlled study. Lancet.
K, Hirasawa R, et al. Effect of Web-based lifestyle 2010;376:10311.
modification on weight control: a meta-analysis. Int 113. NAVIGATOR Study Group. Effect of valsartan on
J Obes (Lond). 2012;36:67585. the incidence of diabetes and cardiovascular events.
102. Kohl LF, Crutzen R, de Vries NK. Online prevention N Engl J Med. 2010;362:147790 (Erratum in: N
aimed at lifestyle behaviors: a systematic review of Engl J Med. 2010;362:1748).
reviews. J Med Internet Res. 2013;15:e146. 114. DPP Research Group. Effects of withdrawal from
doi:10.2196/jmir.2665. metformin on the development of diabetes in the dia-
103. Watson S, Woodside JV, Ware LJ, Hunter SJ, betes prevention program. Diabetes Care. 2003;26:
McGrath A, Cardwell CR, et al. Effect of a web- 97780.
based behavior change program on weight loss and 115. DREAM Trial Investigators. Incidence of diabetes
cardiovascular risk factors in overweight and obese following ramipril or rosiglitazone withdrawal.
adults at high risk of developing cardiovascular dis- Diabetes Care. 2011;34:12659.
ease: randomized controlled trial. J Med Internet 116. Tripathy D, Schwenke DC, Banerji M, Bray GA,
Res. 2015;17(7):e177. doi:10.2196/jmir.3828. Buchanan TA, Clement SC, et al. Diabetes incidence
104. US Department of Agriculture Center for Nutrition and glucose tolerance after termination of piogli-
Policy and Promotion. Dietary guidelines for tazone therapy: results from ACT NOW. J Clin
Americans. Endocrinol Metab. 2016:jc20154202. [Epub ahead
lines.htm. Accessed 5 Apr 2016. of print].
105. Camelon KM, Hdell K, Jmsn PT, Ketonen KJ, 117. Ramachandran A, Snehalatha C, Mary S, Selvam S,
Kohtamki HM, Mkimatilla S, et al. The Plate Model: Kumar CK, Seeli AC, Shetty AS. Pioglitazone does
2 Primary Prevention of Type 2 Diabetes: An Imperative for Developing Countries 31

not enhance the effectiveness of lifestyle modifica- have pre-diabetes and should be considered for met-
tion in preventing conversion of impaired glucose formin therapy. Diabetes Care. 2010;33:4954.
tolerance to diabetes in Asian Indians: results of the 125. Diabetes Prevention Program (DPP) Research
Indian Diabetes Prevention Programme-2 (IDPP-2). Group. Cost associated with the primary prevention
Diabetologia. 2009;52:101926. of type 2 diabetes mellitus in the Diabetes Prevention
118. Greenway FL. Physiological adaptations to weight Program. Diabetes Care. 2003;26:3647.
loss and factors favouring weight regain. Int J Obes 126. Diabetes Prevention Program Research Group.
(Lond). 2015. doi:10.1038/ijo.2015.59. Within-trial cost-effectiveness of lifestyle interven-
119. Nathan DM, Davidson MB, DeFronzo RA, Heine tion or metformin for the primary prevention of type
RJ, Henry RR, Pratley R, Zinman B. Impaired fast- 2 diabetes. Diabetes Care. 2003;26:251823.
ing glucose and impaired glucose tolerance. Diabetes 127. Herman WH, Hoerger TJ, Brandle M, Hicks K,
Care. 2007;30:7539. Sorensen S, Zhang P, Diabetes Prevention Program
120. Twigg SM, Kamp MC, Davis TM, Neylon EK, Flack Research Group, et al. The cost-effectiveness of life-
JR. Australian Diabetes Society and Australian style modification or metformin in preventing type 2
Diabetes Educators Association. Prediabetes: a posi- diabetes in adults with impaired glucose tolerance.
tion statement from the Australian Diabetes Society Ann Intern Med. 2005;142:32332.
and Australian Diabetes Educators Association. 128. Diabetes Prevention Program Research Group. The
Med J Aust. 2007;186:4615. 10-year cost-effectiveness of lifestyle intervention or
121. Indian Health Services. 2008 IHS guidelines for care of metformin for diabetes prevention: an intent-to-treat
adults with prediabetes and/or the metabolic syndrome analysis of the DPP/DPPOS. Diabetes Care. 2012;
in clinical settings. 35:72330.
mon/initiative/DP_Appendices/DPIAN6%20IHS%20 129. Ramachandran A, Snehalatha C, Yamuna A, Mary S,
Guidelines%20PreDiabetes%20Metsyn%20Sept%20 Ping Z. Cost-effectiveness of the interventions in the
2008.pdf. Accessed 3 Apr 2016. primary prevention of diabetes among Asian Indians:
122. Dagogo-Jack S, Askari H, Tykodi G. Glucoregulatory within-trial results of the Indian Diabetes Prevention
physiology in subjects with low-normal, high- Programme (IDPP). Diabetes Care. 2007;30:254852.
normal, or impaired fasting glucose. J Clin 130. Ackermann RT, Marrero DG, Hicks KA, Hoerger
Endocrinol Metab. 2009;94:20316. TJ, Sorensen S, Zhang P, et al. An evaluation of cost
123. Kanat M, Mari A, Norton L, Winnier D, DeFronzo sharing to finance a diet and physical activity inter-
RA, Jenkinson C, Abdul-Ghani MA. Distinct -cell vention to prevent diabetes. Diabetes Care. 2006;29:
defects in impaired fasting glucose and impaired 123741.
glucose tolerance. Diabetes. 2012;61:44753. 131. Dagogo-Jack S. 2015 Presidential address: 75 years
124. Rhee MK, Herrick K, Ziemer DC, Vaccarino V, of battling diabetes-our global challenge. Diabetes
Weintraub WS, Narayan KM, et al. Many Americans Care. 2016;39:39.
Diabetes in Sub-Saharan Africa
Felix Assah and Jean Claude Mbanya

Overview Diabetes prevalence in adults is in general

much higher on islands in sub-Saharan Africa,
Type 2 diabetes1 is a global problem with major compared to the mainland. The highest prevalence
public health and socioeconomic challenges. is found in the Seychelles (17.4 % age-adjusted
From recent estimates of the International comparative prevalence, 17.4 % raw prevalence),
Diabetes Federation (IDF) [1], the number of followed by the island of Reunion (15.8 % age-
adults with diabetes in the world currently stands adjusted, 18.2 % raw) and Comoros (9.9 % age-
at 415 million with a projected rise to 642 million adjusted, 7.5 % raw). Some of Africas most
by 2040. An estimated 14.2 (9.529.4) million populous countries have the highest numbers of
people aged 2079 have diabetes in the sub- people with diabetes, including South Africa (2.3
Saharan Africa (SSA) region, representing a [1.24.6] million), Democratic Republic of
regional prevalence of 2.16.7 %. SSA has the Congo (1.8 [1.52.2] million), Nigeria (1.6 [1.2
highest proportion of undiagnosed cases of dia- 3.8] million) and Ethiopia (1.3 [0.83.5] mil-
betes; over two-thirds (66.7 %) of people with lion). Nearly half of all adults with diabetes in the
diabetes are unaware of their status. The majority region live in these four countries [1].
of people with diabetes (58.8 %) live in cities, Despite the paucity of data from Africa, over
even though the population in the region (61.3 %) the past few decades, diabetes, which was previ-
is predominantly rural. With increasing urbanisa- ously considered to be rare or unknown in rural
tion and population ageing, diabetes will pose an Africa, has emerged as an important non-
even greater threat. It is expected that by 2040 communicable disease (NCD) in the region
there will be 34.2 million adults in the region liv- [24]. Global estimates and projections [1, 57]
ing with diabetes, more than double the number confirm the diabetes epidemic, through the
in 2015 [1]. increasing numbers of people with diabetes and
impaired glucose tolerance (IGT). These reports
not only show that diabetes in adults is a global
This chapter will focus on type 2 diabetes and will be problem but also that populations of developing
referred to as diabetes throughout.
countries, minority groups and disadvantaged
communities in industrialised countries face the
F. Assah, MD, PhD (*) J.C. Mbanya, MD, PhD, FRCP greatest risk. The large increase that is expected
Faculty of Medicine and Biomedical Sciences,
to occur in developing regions of the world has
University of Yaound I, Yaound, Cameroon
e-mail:; principally been attributed to population ageing and urbanisation [1, 5, 7].

Springer International Publishing Switzerland 2017 33

S. Dagogo-Jack (ed.), Diabetes Mellitus in Developing Countries and Underserved Communities,
DOI 10.1007/978-3-319-41559-8_3
34 F. Assah and J.C. Mbanya

Epidemiology of Diabetes in Africa The burden of diabetes in sub-Saharan Africa is

already substantial and continues to increase at a
Historical Perspective very rapid rate. A recent review paper reported that
diabetes in sub-Saharan African countries has
In 1901, Doctor Cook, a medical missionary reached epic proportions. The overall pooled prev-
working in Uganda, reported that diabetes was alence was estimated at 5.7 % for diabetes, 4.5 %
rather uncommon, but very fatal in sub-Saharan for impaired fasting glycaemia and 7.9 % for
Africa [8]. For over half a century after that, dia- impaired glucose tolerance [13]. Diabetes preva-
betes continued to be regarded as a disease of lence in sub-Saharan African countries varies by
affluent societies. Almost no data was available two factors: the first is the level of urbanisation,
as very few studies were carried out on diabetes from 1 % in rural Uganda to 12 % in urban Kenya,
in sub-Saharan Africa until the latter half of the and the second is by ethnicity, from 8 % in
twentieth century. The limited number of early Zimbabwe to 18 % in sub-Saharan African coun-
studies completed between the 1960s and early tries with advanced economies or with significant
1980s reported prevalence rates of mostly <1 %, numbers of Indian subpopulations such as South
except for South Africa with 0.63.6 % and Africa, Kenya and Seychelles [10, 14]. Data from
Ivory Coast with 5.7 % [2]. Doctor Cooks ini- several point prevalence studies in Cameroon show
tial report is no longer accurate today as it is a more than tenfold increase in diabetes prevalence
known that most of the people living with diabe- over a period of one decade [15]; Fig. 3.1.
tes are in developing countries. The burden of Current predictions estimate that the number
diabetes in sub-Saharan Africa is already sub- of people living with diabetes on the continent
stantial and continues to increase at a very rapid will almost double over the coming 20 years [1,
rate. Current predictions estimate that the num- 51, 110]. This, together with the growing inci-
ber of people living with the chronic condition dence of type 2 diabetes in younger age groups
on the continent will almost double over the (including in some obese children even before
next 20 years [1]. puberty), constitutes a serious cause for concern.
In developed countries, most people with diabetes
are above the age of retirement (>60 years), whilst
Current Trends in developing countries, those most frequently
affected are in the middle, productive years of
Sub-Saharan Africa is currently facing a multi- their lives, aged between 35 and 64 [16]. All types
ple disease burden characterised by simultane- of diabetes are on the increase, particularly type 2
ous health challenges from chronic diabetes, and the number of people with diabetes
non-communicable diseases (NCDS), HIV/ will increase by 55 % by 2035 [1] (Fig. 3.2) (IDF
AIDS, other infectious diseases and malnutrition 2015). This underlines the fact that the economic
[9]. The emerging and increasing burden of obe- burden of diabetes in developing countries will be
sity, diabetes and other NCDs is not accompa- amplified by its effect on the working age group.
nied by any marked improvements in control of
infections or undernutrition. Studies have
reported high rates of coexistence of obesity and Risk Factors for Diabetes in Africa [9]
undernutrition in the same communities and
even same households [10, 11]. The current epi- Age, Gender and Family History
demiologic transition in sub-Saharan Africa is
unique and not the same as most Western coun- Similar to data from other parts of the world, stud-
tries. The transition in Africa follows a delayed ies from sub-Saharan Africa (SSA) show that dia-
model [12] which is characterised by high fertil- betes prevalence increases with age (Fig. 3.3)
ity and high but reducing mortality, resulting in a [1821]. Therefore, as living standards improve
double demographic disease pattern. in SSA with a resulting increase in life expectancy,
3 Diabetes in Sub-Saharan Africa 35

Men urban
Women urban
Men rural
Prevalence (%)
Women rural
1994 1998 2003

Fig. 3.1 Prevalence of type 2 diabetes in Cameroon, 19942003 (prevalence standardised to new world population

IDF 2013 2035 Increase

region Millions Millions %
Africa 19.8 41.4 109 %
Middle East and North Africa 34.6 67.9 96 %
South-East Asia 72.1 123 71 %
South and Central America 24.1 38.5 60 %
Western Pacific 138.2 201.8 46 %
North America and Caribbean 36.7 50.4 37 %
Europe 56.3 68.9 22 %
World 381.8 591.9 55 %

Fig. 3.2 The number of people living with diabetes in Africa, EUR Europe, NAC North America and Caribbean,
2013 and projections for 2035 (in millions of cases), with SACA South and Central America, SEA South-East Asia,
projected percent changes (Source of data: IDF Diabetes WP Western Pacific)
Atlas [17]. AFR Africa, MENA Middle East and North

Prevalence (%) estimates of

diabetes by age
(2079 years) and sex

<2 2
46 1
>10 0
20-29 30-39 40-49 50-59 60-69 70-79

* Comparative prevalence

Fig. 3.3 An estimated number of adults with diabetes in sub-Saharan Africa by age group and sex (Source of data: IDF [1])
36 F. Assah and J.C. Mbanya

the prevalence of diabetes is expected to increase Asian Indians [2729] and when compared with a
accordingly. In addition, the age at onset of diabe- population of mixed ancestry (Khoi-East Indian-
tes may shift more towards younger adults due to Europid) [30] (Table 3.1). The prevalence rates in
the effects of urbanisation and westernisation Caucasians in SSA have been quite similar to those
increasing obesity, sedentary lifestyle and other in Europe (610 %), also relatively high [31].
diabetes risk factors in younger age groups. SSA is experiencing the fastest rate of urban-
There is no discernable gender difference in isation worldwide, with an average annual rate of
either the prevalence or distribution of diabetes in change of the urban population of more than 3 %
SSA [22]. This is consistent with observations [34]. Currently, more than one-third of the popu-
from other parts of the world as well as with lation in SSA live in urban areas. It is estimated
global estimates [23]. It must also be noted that a that this could increase to 45 % by 2025, with a
positive family history of diabetes is an indepen- demographic inflection point to be attained by
dent diabetes risk factor [18, 19, 24]. 2035 more urban than rural residents [35]. This
increase in the urban population is influenced by
both the inherent growth of the urban population
Ethnicity, Urbanisation, Migration due to persistent high fertility and longer life
and Diabetes expectancy, and also a massive rural-urban
migration [34]. This rapid urbanisation in SSA
Studies from Tanzania and South Africa have shown has been identified as a major determinant of the
that diabetes prevalence is lower in the indigenous rising burden of diabetes and other cardiovascu-
African community [20, 25, 26] than in migrant lar diseases [12, 23, 36, 37].

Table 3.1 Ethnic differences in the prevalence of diabetes mellitus (D) and impaired glucose tolerance (IGT) and the
effect of migration in Africa
Country Author year Ethnic group Prevalence (%)a
Locality (Reference) D IGT
Tanzania McLarty (1989) [26] Native African 1.1 8.4
Ramaiya (1991) [27] Asian Indian 9.1 16.2
Swai (1990) [28] Asian Indian 7.1 21.5
South Africa
KwaZulu Natal Omar (1993) [25] Native African 5.3 7.7
Omar (1994) [29] Asian Indian 13.0 6.9
Cape Province Levitt (1993) [20] Native African 8.0 7.0
Levitt (1999) [30] Mixed ancestry 10.8 10.2
Sudan Elbagir (1996) [19] Native African 3.4b 2.9b
Elbagir (1998) [18] Mixed Egyptian 10.4 9.8
Nigeria Cooper (1997) [32] Native African 2.0
Caribbean African origin 7.2
United States African origin 10.8
United Kingdom African origin 10.6
Cameroon Rural Mbanya (1999) [33] Native African 0.8 6.4/3.1c
Cameroon Urban Native African 2.0 1.6/4.6c
Caribbean African origin 8.5 16.3/19.6c
United Kingdom African origin 14.6 11.1/14.4c
Age-adjusted prevalence except where indicated
For crude prevalence
3 Diabetes in Sub-Saharan Africa 37

Many studies have clearly demonstrated a first 6 months of urban residence, there was no
positive rural-urban gradient in the prevalence of significant change in their HbA1c level compared
diabetes and its risk factors, particularly obesity. to an age, sex and village-matched nonmigrant
Urban residence is associated with a two- to five- cohort. However, there was an increase in mean
fold increased risk of prevalent diabetes or body mass index (BMI) in the recent migrants
impaired fasting glycaemia (Fig. 3.4) [18, 21, 33, compared to the nonmigrants. More data is
36, 3842]. Most of these studies considered needed to provide answers about societal changes
only the current residence and so may be con- and their influence on human behaviour and car-
founded by the effects of recent rural-urban diovascular risk factors such as diabetes.
migration or vice versa. A study from Cameroon A limitation to the discussion on urban and
[43] that examined both current urban residence rural differences is the lack of a universally
as well as total lifetime exposure to an urban accepted definition for an urban area. Studies
environment found that both lifetime exposure to of rural-urban differences in diabetes have
an urban environment and current residence were used local or national norms to define urban-
independently associated with diabetes. Lifetime ised regions, which may not be similar across
exposure to an urban environment was strongly studies. The existence of geographically sepa-
associated with fasting blood glucose (r = 023; rate residential areas by social class within
P < 0001), with the prevalence of diabetes or IFG urban areas may further confound observed
being higher for individuals with a longer expo- differences between studies. Apart from het-
sure to the urban environment. erogeneity in the definition and attributes of
Most of these studies from SSA are cross- urban area across studies, the degree of
sectional studies, requiring that inferences about urbanisation may influence the magnitude of
causality or direction be made with caution. differences in estimates of diabetes or its risk
Longitudinal studies to provide more robust data factors. A study in Benin [44] that recruited
on direction and magnitude of change are almost participants from rural, semiurban and urban
inexistent. In a prospective study of recent rural- areas observed a positive rural and semiurban
to-urban migrants in Tanzania [42], during the to urban gradient in metabolic syndrome and


Fig. 3.4 Prevalence of 0

diabetes mellitus (D) in









rural [ ] and urban [ ]









communities in community





studies in Africa, using





randomly selected samples

38 F. Assah and J.C. Mbanya

obesity, but not for blood glucose or prevalence generally neglected in terms of healthcare and
of diabetes. The prevalence of high fasting research, hence the scanty nature of data about
blood glucose was significantly lower in urban them. A study from South Africa reported a dia-
dwellers compared to their semiurban or rural betes prevalence of about 5 % among the QwaQwa
counterparts. people [46]. There is currently an increasing
The impact of environmental influence global awareness of the vulnerability of indige-
(migration) in populations of similar genetic ori- nous peoples as a result of their socioeconomic
gin has been confirmed in two studies which disadvantage, limited access to care and margin-
showed that diabetes prevalence was lower in alisation from the majority of the population [17].
native West African populations in Nigeria and From the above data, it appears to be difficult
Cameroon than in West Africanorigin commu- to categorise ethnicity, urbanisation and migra-
nities living abroad, in the Caribbean, United tion as distinct and separate risk factors for diabe-
Kingdom and United States [32, 33] (Table 3.1). tes. Instead, the correlation should be noted and
In a study comparing rural and urban Ghanaians further study undertaken in this area.
in Ghana with Ghanaians in the Netherlands [45]
adjusting for age and educational level, the Diabetes and Obesity
Ghanaians living in the Netherlands were more Published studies indicate a simultaneous rise in
likely to be obese than those in rural Ghana. overweight/obesity and diabetes prevalence in
Currently, there are growing concerns about most SSA countries [47] (Fig. 3.5). Most people
the diabetes burden in indigenous populations with diabetes are obese at the time of diagnosis
because these populations may be experiencing a [48]. Studies conducted in African countries
very rapid lifestyle transition. Active subsistence (Ghana [49, 50], Togo [51], Rwanda [52], South
lifestyle is rapidly giving way to sedentary Africa [53] and Nigeria [69]) have reported sig-
Western lifestyles. Indigenous populations are nificant associations between type 2 diabetes and








a n a i n bia ea ya wi ali ia e n s a o ia e
gol ni an nd oo n n la M tan iqu nio elle fric Tog an bw
An Be tsw uru er am Gui Ke Ma r i b eu ch A n z ba
B am G q au am R ey uth Ta Zim
C E M oz S So

Fig. 3.5 Comparison of overweight/obesity and diabetes prevalence in sub-Saharan Africa (SSA)
3 Diabetes in Sub-Saharan Africa 39

obesity. Many studies of people living with dia- escalate the development of these disorders. This
betes in Africa have consistently reported excess combination of childhood stunting and adulthood
body weight in >50 % of studied samples. obesity has led to the proposition that such indi-
However, there is lack of agreement regarding viduals in their early years acquire the conten-
which of the anthropometric variables used to tious thrifty phenotype such that when they
measure obesity can best predict the develop- rapidly adapt to sedentary lifestyles and high-
ment of diabetes. One case-control study con- calorie diets later in their lives, they will be highly
ducted in a Ghanaian sample [50] suggests that prone to obesity-related disorders [59, 63]. The
measures of central, rather than general, obesity thrifty phenotype hypothesis seems to apply bet-
appear to be significantly related to type 2 diabe- ter for the sub-Saharan African populations
tes in Africans. There is little or no data from rather than the thrifty gene hypothesis (theorises
longitudinal studies with hard endpoints that can that genes derived from times of deprivation may
shed more light on the predictive value of these result in adaptations that have adverse effects in
different measures of obesity in Africans. This times of plenty) [59, 64]. This is because the
means that there are no African-specific cut-off prevalence rates of T2DM and obesity have been
points for estimating obesity risk, hence the con- on the increase since the advent of dramatic
tinued use of Caucasian cut-off points for nutritional/lifestyle changes.
Africans. Further research is needed to establish
cut-off points based on African data [55].
Physical Activity, Diet and Diabetes

Adiposity Available data strongly links the current epide-

miological transition in SSA with rapid urbanisa-
In most studies in sub-Saharan Africa (SSA), tion and westernisation of lifestyle. This changing
adiposity has generally been found to be associ- lifestyle is characterised by decreased levels of
ated with diabetes. These studies have shown physical activity and increased consumption of
that diabetes prevalence is higher with increased energy dense or high-fat diets. Although the sug-
BMI (BMI-specific rates) [24, 26, 32, 56], gestion of lower physical activity in urban areas
waist-hip ratio (WHR-specific rates) [20, 24, compared to rural areas is intuitively plausible,
57] and waist circumference (WC-specific rates) most of the studies on physical activity levels
[24]. This means that overall obesity (BMI) as have used self-reported information, mainly
well as central obesity (WC and WHR) are asso- using physical activity questionnaires which have
ciated with an increased risk for diabetes. not been validated in the population [30, 43, 65
Some researchers have been proposing theories 69]. Few studies have used locally validated
suggesting that some alternate distributions of physical activity questionnaires to demonstrate
body fat (such as in the hips, thighs or legs) could an urbanisation gradient [42, 70] and the
confer lower risk for diabetes. The protective role increased risk of diabetes associated with reduced
of hip circumference as observed in rural South physical activity levels [30].
Africans, confirming recent findings in Australians, Studies documenting temporal trends of phys-
requires further evaluation in Africa [24]. ical activity levels in SSA using comparable
Emerging evidence is demonstrating that pres- methods are lacking. The adoption and use of the
ent day lifestyle habits are not the only contribut- WHO STEPS Global Physical Activity
ing factor to adiposity. The developmental origins Questionnaire for surveillance of habitual physi-
of adult disease suggest that intrauterine growth cal activity by many SSA countries [71] will
retardation and the resulting low birth weight hopefully provide some answers in the near
predispose individuals to metabolic disorders in future about trends in population levels of physi-
their adulthood [5860]; the high prevalence of cal activity over time despite the limitations of
stunting and malnutrition [61, 62] in SSA might self-reported physical activity [68, 69].
40 F. Assah and J.C. Mbanya

Most of the studies which have used objective that the adult population will be decimated by
methods to measure physical activity or energy HIV/AIDS and that few will live long enough to
expenditure are small etiologic studies in specific develop such diseases. However, a South African
populations or questionnaire validation studies study [80], which modelled the impact of HIV/
[6567, 70, 7274] and may not be representative AIDS for the years 1995 and 2010, clearly
of the general population. However, a study in showed that the total number of people with dia-
Cameroon [65] using the doubly labelled water betes and the number of people with diagnosed
method to measure physical activity in free-living diabetes (patient load) will increase regardless of
adults found physical activity energy expenditure the expected impact of HIV/AIDS on population
to be inversely associated with 2-h glucose levels growth rates and any change (no change, increase)
independent of age, sex, adiposity or aerobic fit- in diabetes prevalence. Since this analysis did not
ness. Larger studies in Cameroon and Kenya take into consideration the potential impact of
have used a combined heart rate and motion sen- antiretroviral therapy (ART) on improving sur-
sor to objectively measure physical activity in vival rates or possible increase in diabetes inci-
free-living individuals [39, 75, 76]. These studies dence, it is likely to be a conservative estimate.
demonstrate a significantly higher physical activ- However, the ART of choice in sub-Saharan
ity level in rural compared to urban dwellers, and Africa, with its limited healthcare resources, has
a beneficial association between physical activity been linked with an increase in the numbers of
and abnormal glucose tolerance. people developing prediabetes, and its increased
A high-fat, high-calorie diet is associated use in the treatment of HIV/AIDS is also pro-
with the development of obesity and diabetes. jected to cause adverse metabolic abnormalities
The difficulties of assessing diet and nutritional among patients in SSA [81]. Because of the sheer
data in free-living individuals in epidemiologic numbers of people with HIV/AIDS, this effect
studies pose an even bigger challenge in could directly influence the emerging diabetes
many resource-limited settings. Consequently, crisis in sub-Saharan Africa. Long-term studies
population level dietary data relevant to diabe- to examine the effect of ART on the rising burden
tes and other chronic NCDs in SSA are scanty of diabetes in SSA are needed.
[7779]. A study in South African Blacks
showed a temporal trend of increasing con-
sumption of fat and decreasing carbohydrates Clinical Manifestations
in both urban and rural areas [77]. However,
urbanisation may also lead to a better supply of The presence of atypical forms of diabetes in
fruits and vegetables which improve the micro- sub-Saharan Africa makes it sometimes difficult
nutrient and fibre content of the diet in urban to classify persons living with diabetes based on
compared to rural areas [79]. The notion of established clinical criteria. The disease process
increased consumption of fat as a marker of may involve peripheral resistance to insulin,
urbanisation and westernisation of the African increased hepatic production of glucose and lack
diet is not supported by the findings of a study of insulin production from the pancreas. There
by Mennen et al. which reported the highest are also other causes of diabetes such as genetic
consumption of fat in rural Cameroonians, abnormalities, surgery, drug usage and infectious
compared to urban Cameroonians, Black diseases [82].
Jamaicans and Blacks in Manchester, UK [78]. Diabetes mellitus is defined as a metabolic
disorder caused by different factors characterized
by a chronic high level of blood sugar with distur-
Impact of HIV/AIDS bances to carbohydrate, fat, and protein metabo-
lism resulting from defects in insulin secretion,
In SSA, chronic NCDs such as diabetes receive insulin action, or both [83]. Scientists have
limited attention due in part to misconceptions divided diabetes into three different types: Type 1
3 Diabetes in Sub-Saharan Africa 41

diabetes mellitus (formerly insulin-dependent hope for widespread use in both clinical and
diabetes mellitus IDDM) or type 1 diabetes is research settings even in remote parts of the
also known as juvenile onset diabetes. Type 2 continent.
diabetes mellitus (non-insulin-dependent diabe- Beran et al. surveyed the availability of diag-
tes mellitus (formerly non-insulin-dependent dia- nostic testing tools in a sample of healthcare set-
betes, NIDDM) or type 2 diabetes adult-onset tings in three countries and found that in
diabetes) is found in individuals who are insulin- Mozambique, urine glucose strips were available
resistant and who usually have relative insulin in only 18 % of health facilities surveyed, ketone
deficiency. Gestational diabetes mellitus (GDM), testing strips in 8 % and blood glucose metres in
the third type, is defined as any degree of glucose 21 %, whilst availability in Mali was 54, 43 and
intolerance with onset or first recognition during 13 % and in Zambia 61, 54 and 49 %, respectively
pregnancy. Recently, diabetologists have added a [87, 88]. Low levels of adequate glucose control
fourth category, tropical diabetes, suggested first in those diagnosed with diabetes were reported in
in 1907, but made popular by Hugh Jones (J-type several prevalence studies [89, 90]. Only 27 % of
diabetes) during the mid-1950s by his study of people diagnosed with type 2 diabetes receiving
13 Jamaican patients [84, 85]. However, tropical treatment in a study conducted in Cameroon had
diabetes is less than 1 % of the diabetes cases in adequately controlled glucose levels [89]. Of the
Africa and is thought to be related to malnutri- 99 people with type 1 diabetes in a Tanzanian
tion [31]. survey, only one person achieved suggested glu-
The diagnosis of diabetes usually involves cose targets [90]. None of the 99 people with type
symptoms of diabetes plus either casual plasma 1 diabetes had the ability to monitor their glucose
glucose concentration 200 mg/dl (11.1 mmol/l), levels at home, and hospitals were unable to rou-
a fasting plasma glucose 126 mg/dl (7.0 mmol/l) tinely do this [68]. A regular supply of insulin
or a 2-h post-load glucose 200 mg/dl was unaffordable for many people with diabetes,
(11.1 mmol/l) during an oral glucose tolerance with 1 months insulin supply costing 19.6 days
test (OGTT) conducted as per WHO recommen- wages in Malawi [91] and 25 % of the minimum
dations using a glucose load containing the wage in Tanzania [92]. One Sudanese study
equivalent of 75 g anhydrous glucose dissolved found that 65 % of a familys annual household
in water [48]. In 2010, a WHO expert committee expenditure on health was spent on caring for a
further adopted cut-off points for using glycated diabetic child [93].
haemoglobin (HbA1c) in the diagnosis of diabe- Beran and Yudkin found that state interven-
tes. It concluded that HbA1c can be used as a tions affected insulin price, reporting that an
diagnostic test for diabetes providing that strin- annual supply of insulin cost 5 % of GDP in
gent quality assurance tests are in place and Mozambique, where it was subsidised by the
assays are standardised to criteria aligned to the government, whereas it cost 25 % of GDP in Mali
international reference values, and there are no without subsidies [88]. One study investigated
conditions present which preclude its accurate insulin availability and reported that one in five
measurement. An HbA1c of 6.5 % was therefore hospitals and none out of six health centres sur-
recommended as the cut-off point for diagnosing veyed had a regular insulin supply [87].
diabetes, though a value less than 6.5 % does not
exclude diabetes diagnosed using glucose tests
[86]. The practicality of using HbA1c for diabe- Economic Cost of Diabetes
tes diagnosis in Africa still faces challenges, par-
ticularly the high cost of the test relative to blood Healthcare in most of SSA is almost entirely pri-
glucose measurements and also the high preva- vately purchased, even though the majority of the
lence of haemoglobinopathies such as sickle cell poorest people on earth live in SSA. A recent study
anaemia. As the OGTT is much more demanding by Kirigia et al. [94] clearly demonstrates that the
in terms of time and logistics, HbA1c offers real cost of diabetes care is going to be overwhelming
42 F. Assah and J.C. Mbanya

for the poorest countries of the region. This study insulin. In a study on type 1 diabetes patients in
shows that whilst the direct cost of diabetes per Sudan, Elrayah et al. [93] reported a partial direct
person with diabetes is only a fraction (<25 %) of cost of care for each child with type 1 diabetes of
the GNI per capita for the 12 richest countries, the USD 283, in a country with a GDP per capita of
direct cost for the 34 poorest countries of the USD 300 and a per capita government expendi-
region is 125 % of their GNI per capita (Fig. 3.6). ture on health of USD 300.
For these poorest countries, the total cost (direct These data indicate undeniably that the cost of
and indirect costs) of diabetes per person with dia- diabetes care constitutes a huge burden for
betes is more than double the GNI per capita. SSA. In the absence of a publicly funded health-
The few studies which have examined the cost care system, these costs are borne almost entirely
of diabetes care (usually in small samples) in the by individuals individuals who are among the
region confirm the above estimates. Akoussou- poorest people in the world. In this context, poor
Zinsou and Amedegnato reported in 2001 that the disease prognosis, with high morbidity and mor-
direct cost of diabetes care at a teaching hospital tality seem to be the unavoidable outcome.
in Togo was USD 342 and USD 110 per person
for complicated and uncomplicated diabetes
patients, respectively. The estimated GNI per Healthcare Access for Diabetes
capita for Togo at the time was approximately and its Complications
USD 385. Chale et al. [95] reported an average
annual direct cost of diabetes care in Tanzania in Many SSA countries have made significant
19891990 of USD 287 for a patient requiring efforts in initiating and improving care for diabe-
insulin and USD 103 for a patient not requiring tes and other chronic diseases. However, two-


Average GNI per capita

Amount (thousands of International dollars)

Average direct cost per person with diabetes

Average indirect cost per person with diabetes


Average total cost per person with diabetes


Group 1 Group 2 Group 3

Fig. 3.6 Direct, indirect and total cost of diabetes per ing to the average GNI per capita of the countries (Adapted
person with diabetes compared to the average GNI per from Kirigia et al. [94])
capita in countries of WHO Africa region grouped accord-
3 Diabetes in Sub-Saharan Africa 43

thirds of people with diabetes in low- and Control of blood glucose and blood pressure is
middle-income countries have poor diabetes con- very poor.
trol because of inadequate access to care oral Referral systems are inadequate.
antidiabetic agents and insulin are not available Patient education is almost nonexistent.
at all times in many parts of the developing world Organisation of services is generally poor.
[96]. A few countries have now established suc- Better record keeping will assist improve-
cessful National Diabetes Programmes and ments in care.
guidelines for the management of diabetes [97
99]. For most countries, there is still much Therefore, even when diabetes care pro-
improvement required. Limited financial grammes are introduced, a lot will still need to be
resources are a major handicap even with the done to enable them meet their objective of
availability of scientific information and political reducing morbidity and mortality linked to diabe-
will both of which are still largely lacking in tes and its complications. A few of the drawbacks
many SSA countries. above relate to inadequate healthcare staff levels
It is possible to set up effective diabetes as well as patient-related factors. Patient educa-
care at the peripheral level using basic tion and empowerment in the form of a peer sup-
resources in terms of personnel and equipment. port system could offer some hope of better
Nurse-led care including education has been diabetes control by improving compliance and
shown to be successful in resource-limited set- optimising self-care practices.
tings over an 18-month period [100]. It would
however be a real challenge to translate such
experiments over a larger scale and a much Advocacy and Policy for Diabetes
longer period. Work carried out by the Control and Prevention
International Insulin Foundation [101] identi-
fied key areas to be addressed in the fight The adoption of a World Diabetes Day
against diabetes in SSA, encompassing policy Resolution by the United Nations General
improvement, better organisation and delivery Assembly [103] is acknowledgement of the
of care and patient education. Whiting et al. fact that diabetes is a real and imminent threat
[102] reviewed challenges to the delivery of to social and economic development globally.
diabetes care in a few countries which had The UN resolution calls for member states to
reported data on diabetes healthcare delivery, develop national policies for the prevention,
and could still identify the following barriers treatment and care of diabetes in line with the
to adequate diabetes care: sustainable development of their healthcare
systems, taking into account the internation-
Patient attendance is poor (there are many rea- ally agreed development goals, including the
sons for this and more work is required to Millennium Development Goals. In SSA spe-
explore them). cifically, the IDF Africa Region, the World
Consultation times are very short, leaving lit- Health Organization (WHO)-AFRO and the
tle or no time for patient education. African Union have jointly issued a Diabetes
Staff levels are inadequate and more use could Declaration and Strategy for Africa [104]. The
be made of trained nurses and other health declaration is a call to action for governments
workers. of African countries and all partners and
Staff training is limited, and continuing edu- stakeholders in diabetes to prevent diabetes
cation or in-service training, especially of and related non-communicable diseases and
lower cadres, is needed. to improve quality of life and reduce morbid-
Complications are not monitored or evaluated ity and premature mortality from diabetes
in a systematic manner. (Panel).
44 F. Assah and J.C. Mbanya

Future Perspectives of Diabetes

The IDF-Africa, the WHO-AFRO and Prevention and Control
the Africa Union call on governments of
African countries, non-government organ- Diabetes is perhaps the index case of the general
isations, international donor agencies, problem of non-communicable disease health-
industry, health care providers and all part- care delivery in developing countries. There have
ners in diabetes to ensure: been scattered reports [100, 105109] of success-
ful attempts to improve diabetes care delivery
Adequate, appropriate and affordable and outcome, but sadly these have largely been
medications and supplies for people liv- initiated either by local hospitals or by external
ing with diabetes funding and support. The lessons appear to be
Earlier detection and optimal quality of that real improvements in diabetes care and
care for diabetes outcome in Africa are achievable. However,
Effective efforts to create healthier envi- although external support, local health facilities,
ronments and prevent diabetes support groups and diabetes associations all have
The identification and dissemination of a role to play, national government health depart-
information, education and communica- ments need to take the responsibility of instigat-
tion to empower people with diabetes to ing widespread permanent change and
access appropriate diabetes services and improvement. The costs need not be great, as
improve self-care patient education (one of the least expensive of
Equitable access to care and prevention diabetes treatments) has been shown to be a
services for people with or at risk of major and effective part of all the currently
diabetes described care delivery packages [100, 105109].
Awareness of diabetes in the community The integration of traditional healers should also
and among health care providers be considered as part of these reforms, since, to
A truly integrated approach which uti- the everyday African, they are very much a part
lises the whole health workforce to of illness management.
address infectious and non-
communicable diseases simultaneously
Government commitment to reducing References
the personal and public health burden of
diabetes 1. International Diabetes Federation. IDF diabetes atlas.
Partnership and collaboration within Brussels, Belgium: International Diabetes Federation;
7th ed. 2015.
and between government sectors, pri-
2. McLarty DG, Pollitt C, Swai AB. Diabetes in Africa.
vate sectors, non-government organisa- Diabet Med. 1990;7(8):67084.
tions and communities to create 3. Motala AA, Omar MA, Pirie FJ. Diabetes in Africa.
community and workplace environ- Epidemiology of type 1 and type 2 diabetes in Africa.
J Cardiovasc Risk. 2003;10(2):7783.
ments that promote better health.
4. Levitt NS. Diabetes in Africa: epidemiology, man-
agement and healthcare challenges. Heart. 2008;
5. International Diabetes Federation. IDF diabetes atlas.
Brussels, Belgium: International Diabetes Federation;
The advocacy for diabetes prevention and
3rd ed. 2008.
control in SSA is largely based on general prin- 6. King H, Aubert RE, Herman WH. Global burden
ciples that are generally accepted. There is very of diabetes, 19952025: prevalence, numerical esti-
little evidence available from local data to sub- mates, and projections. Diabetes Care. 1998;21(9):
stantiate the implementation of prevention or
7. Wild S, Roglic G, Green A, Sicree R, King
control interventions. H. Global prevalence of diabetes: estimates for the
3 Diabetes in Sub-Saharan Africa 45

year 2000 and projections for 2030. Diabetes Care. 23. International Diabetes Federation. IDF diabetes atlas.
2004;27:104753. Brussels, Belgium: International Diabetes Federation;
8. Cook A. Notes on the diseases met with in Uganda, 4th ed. 2009.
Central Africa. J Trop Med. 1901;4:1758. 24. Motala AA, Esterhuizen T, Gouws E, Pirie FJ, Omar
9. Mbanya JC, Motala AA, Sobngwi E, Assah FK, MA. Diabetes and other disorders of glycemia in a
Enoru ST. Diabetes in sub-Saharan Africa. Lancet. rural South African community: prevalence and
2010;375(9733):225466. associated risk factors. Diabetes Care. 2008;31(9):
10. Hall V, Thomsen RW, Henriksen O, Lohse N. Diabetes 17838.
in Sub-Saharan Africa 19992011: epidemiology and 25. Omar MA, Seedat MA, Motala AA, Dyer RB, Becker
public health implications. A systematic review. BMC P. The prevalence of diabetes mellitus and impaired
Public Health. 2011;11:564. glucose tolerance in a group of urban South African
11. Kimani-Murage EW, Muthuri SK, Oti SO, Mutua blacks. S Afr Med J. 1993;83(9):6413.
MK, Van de V, Kyobutungi C. Evidence of a dou- 26. McLarty DG, Swai AB, Kitange HM, Masuki G,
ble burden of malnutrition in urban poor settings in Mtinangi BL, Kilima PM, et al. Prevalence of diabe-
Nairobi, Kenya. PLoS One. 2015;10(6):e0129943. tes and impaired glucose tolerance in rural Tanzania.
12. Omran AR. The epidemiologic transition: a theory Lancet. 1989;1(8643):8715.
of the epidemiology of population change. 1971. 27. Ramaiya KL, Swai AB, McLarty DG, Alberti
Milbank Q. 2005;83:73157. KG. Impaired glucose tolerance and diabetes mellitus
13. Hilawe EH, Yatsuya H, Kawaguchi L, Aoyama in Hindu Indian immigrants in Dar es Salaam. Diabet
A. Differences by sex in the prevalence of diabetes Med. 1991;8(8):73844.
mellitus, impaired fasting glycaemia and impaired 28. Swai AB, McLarty DG, Sherrif F, Chuwa LM, Maro
glucose tolerance in sub-Saharan Africa: a systematic E, Lukmanji Z, et al. Diabetes and impaired glu-
review and meta-analysis. Bull World Health Organ. cose tolerance in an Asian community in Tanzania.
2013;91(9):671682D. Diabetes Res Clin Pract. 1990;8(3):22734.
14. Bovet P, Gabriel A, Shamlaye C, Paccaud F. Diabetes in 29. Omar MA, Seedat MA, Dyer RB, Motala AA, Knight
Africa: the situation in the Seychelles. Heart. 2009;95(6): LT, Becker PJ. South African Indians show a high
5067. prevalence of NIDDM and bimodality in plasma
15. Mbanya JC, Kengne AP, Assah F. Diabetes care in glucose distribution patterns. Diabetes Care. 1994;
Africa. Lancet. 2006;368(9548):16289. 17(1):703.
16. Tabish SA. Is diabetes becoming the biggest epidemic 30. Levitt NS, Steyn K, Lambert EV, Reagon G,
of the twenty-first century? Int J Health Sci (Qassim). Lombard CJ, Fourie JM, et al. Modifiable risk
2007;1(2):VVIII. factors for type 2 diabetes mellitus in a peri-urban
17. International Diabetes Federation. IDF diabetes atlas. community in South Africa. Diabet Med. 1999;16:
Brussels, Belgium: International Diabetes Federation; 94650.
6th ed. 2013. 31. Sobngwi E, Mauvais-Jarvis F, Vexiau P, Mbanya JC,
18. Elbagir MN, Eltom MA, Elmahadi EM, Kadam IM, Gautier JF. Diabetes in Africans part 1: epidemi-
Berne C. A high prevalence of diabetes mellitus ology and clinical specificities. Diabetes Metab.
and impaired glucose tolerance in the Danagla com- 2001;27:62834.
munity in northern Sudan. Diabet Med. 1998;15(2): 32. Cooper RS, Rotimi CN, Kaufman JS, Owoaje EE,
1649. Fraser H, Forrester T, et al. Prevalence of NIDDM
19. Elbagir MN, Eltom MA, Elmahadi EM, Kadam IM, among populations of the African diaspora. Diabetes
Berne C. A population-based study of the prevalence Care. 1997;20:3438.
of diabetes and impaired glucose tolerance in adults 33. Mbanya JCN, Cruickshank JK, Forrester T, Balkau B,
in Northern Sudan. Diabetes Care. 1996;19(10): Ngogang JY, Riste L, et al. Standardised comparison
11268. of glucose intolerance in West African-origin popu-
20. Levitt NS, Katzenellenbogen JM, Bradshaw D, lations of rural and urban Cameroon, Jamaica, and
Hoffman MN, Bonnici F. The prevalence and identi- Caribbean migrants to Britain. Diabetes Care. 1999;22:
fication of risk factors for NIDDM in urban Africans 43440.
in Cape Town, South Africa. Diabetes Care. 1993; 34. United Nations Population Fund. State of world popu-
16(4):6017. lation 2007: unleashing the potential of urban growth.
21. Balde NM, Diallo I, Balde MD, Barry IS, Kaba L, 35. Population Division of the Department of Economic
Diallo MM, et al. Diabetes and impaired fasting glu- and Social Affairs of the United Nations Secretariat.
cose in rural and urban populations in Futa Jallon World urbanization prospects: the 2007 revision. 2008
(Guinea): prevalence and associated risk factors. 36. Mayosi BM, Flisher AJ, Lalloo UG, Sitas F, Tollman
Diabetes Metab. 2007;33(2):11420. SM, Bradshaw D. The burden of non-communicable
22. Motala AA, Omar MAK, Pirie FJ. Epidemiology diseases in South Africa. Lancet. 2009;374:93447.
of diabetes in Africa. In: Ekoe J-M, Rewers M, 37. Popkin BM, Gordon-Larsen P. The nutrition transi-
Williams R, Zimmet P, eds. The epidemiology of tion: worldwide obesity dynamics and their determi-
diabetes mellitus (2nd edn). Chichester: Wiley, 2008; nants. Int J Obes Relat Metab Disord. 2004;28 Suppl
13346. 3:S29.
46 F. Assah and J.C. Mbanya

38. Aspray TJ, Mugusi F, Rashid S, Whiting D, Edwards 53. Erasmus RT, Blanco BE, Okesina AB, Gqweta Z,
R, Alberti KG, et al. Rural and urban differences in Matsha T. Assessment of glycaemic control in stable
diabetes prevalence in Tanzania: the role of obesity, type 2 black South African diabetics attending a peri-
physical inactivity and urban living. Trans R Soc Trop urban clinic. Postgrad Med J. 1999;75(888):6036.
Med Hyg. 2000;94:63744. 54. Oguntibeju OO, Odunaiya N, Oladipo B, Truter
39. Christensen DL, Friis H, Mwaniki DL, Kilonzo B, EJ. Health behaviour and quality of life of patients
Tetens I, Boit MK, et al. Prevalence of glucose intol- with type 2 diabetes attending selected hospitals in
erance and associated risk factors in rural and urban South Western Nigeria. West Indian Med J. 2012;
populations of different ethnic groups in Kenya. 61(6):61926.
Diabetes Res Clin Pract. 2009;84:30310. 55. Mbanya JC, Assah FK, Saji J, Atanga EN. Obesity
40. Mbanya JC, Ngogang J, Salah JN, Minkoulou E, and type 2 diabetes in Sub-Saharan Africa. Curr Diab
Balkau B. Prevalence of NIDDM and impaired glu- Rep. 2014;14(7):501.
cose tolerance in a rural and an urban population in 56. Fisch A, Pichard E, Prazuck T, Leblanc H, Sidibe Y,
Cameroon. Diabetologia. 1997;40:8249. Brucker G. Prevalence and risk factors of diabetes mel-
41. Mollentze WF, Moore AJ, Steyn AF, Joubert G, Steyn litus in the rural region of Mali (West Africa): a practi-
K, Oosthuizen GM, et al. Coronary heart disease risk cal approach. Diabetologia. 1987;30(11):85962.
factors in a rural and urban orange free state black 57. Fisch A, Pichard E, Prazuck T, Leblanc H, Cisse
population. S Afr Med J. 1995;85:906. H. Suitability of reflectometry in epidemiology of dia-
42. Unwin N, McLarty D, Machibya H, Aspray T, Tamin betes mellitus in Africans. Lancet. 1990;335(8690):
B, Carlin L, et al. Changes in blood pressure and lipids 6612.
associated with rural to urban migration in Tanzania. 58. Prentice AM, Moore SE. Early programming of adult
J Hum Hypertens. 2006;20:7046. diseases in resource poor countries. Arch Dis Child.
43. Sobngwi E, Mbanya JC, Unwin NC, Porcher R, 2005;90(4):42932.
Kengne AP, Fezeu L, et al. Exposure over the life 59. Joffe B, Zimmet P. The thrifty genotype in type 2 dia-
course to an urban environment and its relation with betes: an unfinished symphony moving to its finale?
obesity, diabetes, and hypertension in rural and urban Endocrine. 1998;9(2):13941.
Cameroon. Int J Epidemiol. 2004;33:76976. 60. Pettitt DJ, Aleck KA, Baird HR, Carraher MJ,
44. Ntandou G, Delisle H, Agueh V, Fayomi B. Bennett PH, Knowler WC. Congenital susceptibil-
Abdominal obesity explains the positive rural-urban ity to NIDDM. Role of intrauterine environment.
gradient in the prevalence of the metabolic syndrome Diabetes. 1988;37(5):6228.
in Benin, West Africa. Nutr Res. 2009;29:1809. 61. Martorell R, Kettel Khan L, Hughes ML, Grummer-
45. Agyemang C, Owusu-Dabo E, de Jonge A, Martins Strawn LM. Overweight and obesity in preschool
D, Ogedegbe G, Stronks K. Overweight and obesity children from developing countries. Int J Obes Relat
among Ghanaian residents in The Netherlands: how Metab Disord. 2000;24:95967. (8): 959967.
do they weigh against their urban and rural counter- 62. de Onis M, Blossner M. Prevalence and trends of
parts in Ghana? Public Health Nutr. 2009;12:90916. overweight among preschool children in developing
46. Mollentze WF, Osthuizen GM, et al. The prevalence countries. Am J Clin Nutr. 2000;72(4):10329.
of diabetes mellitus in two South African black popu- 63. Hossain P, Kawar B, El NM. Obesity and diabetes in
lations. Diabetologia. 1992;5:511. 1992. the developing world a growing challenge. N Engl
47. World Health Organization. Global status report on J Med. 2007;356:2135.
noncommunicable diseases 2010. Geneva: World 64. Levitt NS, Lambert EV. The foetal origins of the
Health Organization; 2011. 2011. metabolic syndrome a South African perspective.
48. American Diabetes Association. Position statement: Cardiovasc J S Afr. 2002;13(4):17980.
diagnosis and classification of diabetes mellitus. 65. Assah FK, Ekelund U, Brage S, Mbanya JC,
Diabetes Care. 2004;27(1):s510. Wareham NJ. Free-living physical activity energy
49. Danquah I, Bedu-Addo G, Terpe KJ, Micah F, expenditure is strongly related to glucose intoler-
Amoako YA, Awuku YA, et al. Diabetes mellitus type ance in Cameroonian adults independently of obesity.
2 in urban Ghana: characteristics and associated fac- Diabetes Care. 2009;32(2):3679.
tors. BMC Public Health. 2012;12:210. 66. Cole AH, Ogbe JO. Energy intake, expenditure and
50. Frank LK, Heraclides A, Danquah I, Bedu-Addo G, pattern of daily activity of Nigerian male students. Br
Mockenhaupt FP, Schulze MB. Measures of gen- J Nutr. 1987;58:35767.
eral and central obesity and risk of type 2 diabetes 67. Luke A, Dugas LR, Ebersole K, Durazo-Arvizu
in a Ghanaian population. Trop Med Int Health. RA, Cao G, Schoeller DA, et al. Energy expenditure
2013;18(2):14151. does not predict weight change in either Nigerian or
51. Barruet R, Gbadoe AD. Type 2 diabetes mellitus in African-American women. Am J Clin Nutr. 2009;89:
children in black Africa: description of first five cases 16976.
in Togo. Med Trop (Mars). 2006;66(5):4813. 68. Neilson HK, Robson PJ, Friedenreich CM, Csizmadi
52. Buresi D. Clinical study of diabetes mellitus in hospi- I. Estimating activity energy expenditure: how valid
tal practice in Northern Rwanda (apropos of 86 case are physical activity questionnaires? Am J Clin Nutr.
reports). Med Trop (Mars). 1988;48(3):22935. 2008;87:27991.
3 Diabetes in Sub-Saharan Africa 47

69. Wareham NJ, Rennie KL. The assessment of physical 83. World Health Organisation. Definition, diagnosis and
activity in individuals and populations: why try to be classification of diabetes mellitus and its complica-
more precise about how physical activity is assessed? tions. WHO/NCD/NCS/99.2. Geneva: World Health
Int J Obes Relat Metab Disord. 1998;22 Suppl 2: Organization; 1999.
S308. 84. McMillan DE. Tropical malnutrition diabetes.
70. Singh J, Prentice AM, Diaz E, Coward WA, Ashford Diabetologia. 1986;29(2):1278.
J, Sawyer M, et al. Energy expenditure of Gambian 85. Ducorps M, Ndong W, Jupkwo B, Belmejdoub G,
women during peak agricultural activity measured by Poirier JM, Mayaudon H, et al. Epidemiological
the doubly-labelled water method. Br J Nutr. 1989;62: aspects of diabetes in Cameroon: what is the role of
31529. tropical diabetes? Diabetes Metab. 1997;23(1):617.
71. World Health Organization. STEPwise approach 86. World Health Organization. Use of glycated hemo-
to chronic disease risk factor surveillance. Geneva: globin (HbA1c) in the diagnosis of diabetes mellitus.
World Health Organization; 2009. Geneva: World Health Organization; 2010.
72. Heini AF, Minghelli G, Diaz E, Prentice AM, Schutz 87. Beran D, Yudkin JS, de Court. Access to care for
Y. Free-living energy expenditure assessed by two dif- patients with insulin-requiring diabetes in developing
ferent methods in rural Gambian men. Eur J Clin Nutr. countries: case studies of Mozambique and Zambia.
1996;50:2849. Diabetes Care. 2005;28(9):213640.
73. Huss-Ashmore R, Goodman JL, Sibiya TE, Stein 88. Beran D, Yudkin JS. Looking beyond the issue of
TP. Energy expenditure of young Swazi women as access to insulin: what is needed for proper diabe-
measured by the doubly-labelled water method. Eur tes care in resource poor settings. Diabetes Res Clin
J Clin Nutr. 1989;43:73748. Pract. 2010;88(3):21721.
74. Kolbe-Alexander TL, Lambert EV, Harkins JB, 89. HoPiT Research Group. Cameroon burden of diabetes
Ekelund U. Comparison of two methods of measur- project (Cambod): baseline survey report. Yaound:
ing physical activity in South African older adults. HoPiT Research Group; 2004.
J Aging Phys Act. 2006;14:98114. 90. Majaliwa ES, Munubhi E, Ramaiya K, Mpembeni R,
75. Dirk Lund Christensen. Glucose intolerance in rela- Sanyiwa A, Mohn A, et al. Survey on acute and chronic
tion to body composition, physical activity and other complications in children and adolescents with type
predictors among the Luo, Kamba and Maasai of 1 diabetes at Muhimbili National Hospital in Dar es
Kenya. PhD Thesis, University of Copenhagen. 2008. Salaam, Tanzania. Diabetes Care. 2007;30(9):218792.
76. Assah FK. Physical activity, cardiorespiratory fit- 91. Mendis S, Fukino K, Cameron A, Laing R, Filipe Jr
ness and metabolic health in Cameroon. PhD Thesis, A, Khatib O, et al. The availability and affordability
University of Cambridge. 2009. of selected essential medicines for chronic diseases
77. Bourne LT, Lambert EV, Steyn K. Where does the in six low- and middle-income countries. Bull World
black population of South Africa stand on the nutri- Health Organ. 2007;85(4):27988.
tion transition? Public Health Nutr. 2002;5:15762. 92. Neuhann HF, Warter-Neuhann C, Lyaruu I, Msuya
78. Mennen LI, Jackson M, Sharma S, Mbanya JCN, L. Diabetes care in Kilimanjaro region: clinical pre-
Cade J, Walker S, et al. Habitual diet in four popula- sentation and problems of patients of the diabetes
tions of African origin: a descriptive paper on nutri- clinic at the regional referral hospital - an inven-
ent intakes in rural and urban Cameroon, Jamaica and tory before structured intervention. Diabet Med.
Caribbean migrants in Britain. Public Health Nutr. 2002;19(6):50913.
2001;4:76572. 93. Elrayah H, Eltom M, Bedri A, Belal A, Rosling H,
79. Vorster HH, Venter CS, Wissing MP, Margetts Ostenson CG. Economic burden on families of child-
BM. The nutrition and health transition in the North hood type 1 diabetes in urban Sudan. Diabetes Res
West Province of South Africa: a review of the Clin Pract. 2005;70(2):15965.
THUSA (Transition and Health during Urbanisation 94. Kirigia JM, Sambo HB, Sambo LG, Barry
of South Africans) study. Public Health Nutr. 2005;8: SP. Economic burden of diabetes mellitus in the
48090. WHO African region. BMC Int Health Hum Rights.
80. Levitt NS, Bradshaw D. The impact of HIV/AIDS on 2009;9:6.
type 2 diabetes prevalence and diabetes healthcare 95. Chale SS, Swai AB, Mujinja PG, McLarty DG. Must
needs in South Africa: projections for 2010. Diabet diabetes be a fatal disease in Africa? Study of costs
Med. 2006;23(1):1034. of treatment. BMJ Clin Res Ed. 1992;304(6836):
81. Young F, Critchley JA, Johnstone LK, Unwin NC. A 12158.
review of co-morbidity between infectious and 96. Deeb LC, Tan MH, Alberti KG. Insulin availability
chronic disease in Sub-Saharan Africa: TB and dia- among International Diabetes Federation member
betes mellitus, HIV and metabolic syndrome, and the associations. Report of the Task Force on Insulin
impact of globalization. Glob Health. 2009;5:9. Distribution. Diabetes Care. 1994;17(3):2203.
82. Idemyor V. Diabetes in sub-Saharan Africa: health- 97. Njamnshi A, Hiag AB, Mbanya J-C. From research
care perspectives, challenges, and the economic to policy: the development of a national diabetes pro-
burden of disease. J Natl Med Assoc. 2010;102(7): gramme in Cameroon. Diabetes Voice. 2006;51(3):
6503. 1821.
48 F. Assah and J.C. Mbanya

98. Ramaiya K. Tanzania and diabetes a model for 106. Acheampong JW, Boateng KA, Eghan BA, Story
developing countries? BMJ. 2005;330(7492):679. P, Parry EHO, Tomlinson S. The impact of diabe-
99. Daniels A, Biesma R, Otten J, Levitt NS, Steyn K, tes nurses in the Komfo Anokye Teaching Hospital,
Martell R, et al. Ambivalence of primary healthcare Ghana. Diabetes Int. 2000;10:8193.
professionals towards the South African guidelines 107. Mamo Y, Seid E, Adams S, Gardiner A, Parry E. A
for hypertension and diabetes. S Afr Med J. 2000; primary healthcare approach to the management of
90(12):120611. chronic disease in Ethiopia: an example for other
100. Gill GV, Price C, Shandu D, Dedicoat M, Wilkinson countries. Clin Med (Lond). 2007;7(3):22831.
D. An effective system of nurse-led diabetes care 108. Coleman R, Gill G, Wilkinson D. Noncommunicable
in rural Africa. Diabet Med. 2008;25(5):60611. disease management in resource-poor settings: a
101. Beran D, Yudkin JS. Diabetes care in sub-Saharan primary care model from rural South Africa. Bull
Africa. Lancet. 2006;368(9548):168995. World Health Organ. 1998;76(6):63340.
102. Whiting DR, Hayes L, Unwin NC. Diabetes in 109. Windus DW, Ladenson JH, Merrins CK, Seyoum
Africa. Challenges to health care for diabetes in M, Windus D, Morin S, et al. Impact of a multidis-
Africa. J Cardiovasc Risk. 2003;10(2):10310. ciplinary intervention for diabetes in Eritrea. Clin
103. United Nations. World diabetes day. A/RES/61/225. Chem. 2007;53(11):19549.
12-20-2006. 110. Otaigbe BE, Imafidon EE. Type 2 diabetes mellitus
104. IDF Africa. Diabetes declaration and strategy for in a Nigerian child: a case report. Afr Health Sci.
Africa. International Diabetes Federation; 2006. 2011;11(3):4546.
105. Huddle KR, Gill GV. Reducing acute hyperglycae-
mic mortality in African diabetic patients. Diabet
Med. 1989;6(1):646.
Type 2 Diabetes in the Middle East
and North Africa (MENA) 4
Yasmin Khan and Osama Hamdy

Introduction lion) [1]. The rapid rise is due to a multitude of

factors, including economic and demographic
The Middle East and North Africa (MENA) changes over the last few decades that have led
region encompasses approximately 22 countries to a decrease in physical activity and rise in obe-
and comprises 6 % of the worlds population. It is sity. The MENA region has among the highest
a geographic area with rich history and linguistic, obesity rates in the world (Fig. 4.3).
ethnic, and cultural diversity. According to WHO Diabetes is associated with early mortality
statistics from 2010, Egyptians account for more and increased risk for microvascular and macro-
than one-third of the population in this region, vascular complications. Approximately 40 % of
followed by North Africans and Arabs from the patients with diabetes have chronic kidney dis-
Gulf (Fig. 4.1). The MENA region has one of the ease, and almost 6070 % of patients with dia-
highest prevalence rates of diabetes in the world betes have mild to severe forms of nervous
(11 %) (Fig. 4.2). system damage. In addition, patients with diabe-
Diabetes is a global pandemic affecting more tes are two to four times more likely to have
than 415 million adults worldwide. This number fatal or nonfatal coronary events or stroke.
is expected to surge to 642 million by 2040 [1]. Almost 7080 % of patients with diabetes die
According to the International Diabetes from one of these two conditions. The American
Federation (IDF), over 35 million people in Heart Association considers diabetes to be one
MENA have diabetes. Saudi Arabia leads the of the six major controllable risk factors for car-
MENA region with the highest prevalence of diovascular disease. Researchers also consider
diabetes (23.9 %), and Egypt is the country with diabetes as a risk equivalent to having a prior
the largest number of diabetes patients (7.5 mil- heart attack.
In addition to being a major public health
problem, diabetes is also associated with signifi-
Y. Khan, MD, MPH (*) cant healthcare costs. Worldwide, the global cost
Adult Diabetes Section, Joslin Diabetes Center, of diabetes continues to soar and is now estimated
Boston, MA, USA
to be 825 billion dollars per year [2]. The IDF
estimates that in 2015, countries in the Middle
O. Hamdy, MD, PhD, FACE
Joslin Diabetes Center, Harvard Medical School,
East spent over 17 billion dollars on diabetes care.
Boston, MA, USA This figure accounts for only 2.5 % of global
e-mail: spending on the disease [1]. Healthcare expenditure

Springer International Publishing Switzerland 2017 49

S. Dagogo-Jack (ed.), Diabetes Mellitus in Developing Countries and Underserved Communities,
DOI 10.1007/978-3-319-41559-8_4
50 Y. Khan and O. Hamdy

Fig. 4.1 Ethnicity of the 26.7 (1%) (3%)

MENA region (Source: (8%)
2010 WHO statistics) lestinians jordanians
Sudaneses 108.4 (33%)

Gulf Arabs 58.9


North Africans 83.6

Fig. 4.2 Comparative

prevalence of diabetes in
MENA (Source: IDF
Diabetes Atlas)

<5 10-12

5-8 12-15
8-10 >15

*comparative prevalence

varies greatly across the region due to economic diabetes to avoid major spending on healthcare in
inequalities. In Egypt, a cost analysis from 2010 the coming years.
estimated the economic impact of diabetes to be
$1.29 billion. This number excluded the cost
associated with prediabetes and also the cost Epidemiology of Diabetes in MENA
related to loss of productivity. This figure, adjusted
for inflation, can be expected to double by the The MENA region can be described as a global
year 2030. According to the IDF, in Egypt the cur- hotspot for diabetes. According to the IDF, approx-
rent spending on diabetes is among the lowest in imately 37 million adults aged 2079 are living
the MENA at $116 per patient per year (16 % of with diabetes across the region. By 2040, this num-
total healthcare expenditure). Countries with the ber is expected to rise to 72 million. Of particular
highest spending per person with diabetes are concern is that over 40 % of individuals with diabe-
Qatar, Kuwait, UAE, and Bahrain, with a range of tes are undiagnosed and are at increased risk for
$1,0002,000 per year. This is still lower than developing complications from diabetes. Moreover,
spending for developed countries, which usually in 2015 there were 342,000 deaths related to diabe-
ranges from $2,000 to 7,000 per patient per year tes. A further 30.2 million people in the region, or
[1]. The rising epidemic will continue to strain the 7.8 % of the adult population, are estimated to have
economies of the MENA region, and health impaired glucose tolerance and are therefore at
authorities should urgently address the problem of high risk of developing diabetes in the future [1].
4 Type 2 Diabetes in the Middle East and North Africa (MENA) 51


Frequency of BMI (%) Overweight (>25 to <30 kg/m2) Obese (> 30 kg/m2)









pe st

As ast
ric h

As uth
st e

ric h
ro th

Af out

Ea ddl

ro t

Af ort
ro e

Eu Eas

er in
Eu ou




Eu th W



Am Lat





SOURCE: American Heart Association


Frequency of BMI (%) Overweight (>25 to <30 kg/m2) Obese (> 30 kg/m2)









pe st

As ast
ric h

As uth
st e

ric h
ro th

Af out

Ea ddl

ro t

Af ort
ro e

Eu Eas

er in
Eu ou




Eu th W



Am Lat





SOURCE: American Heart Association

Fig. 4.3 Global prevalence of obesity (Source: AHA)

52 Y. Khan and O. Hamdy

Prevalence of Diabetes individuals in Egypt have diabetes and another 2.2

million have prediabetes. Furthermore, reports
Diabetes prevalence varies greatly across the indicate that 43 % of patients with diabetes and
region. The highest prevalence is seen in the Gulf most patients with prediabetes in Egypt are likely
countries, with Saudi Arabia, Kuwait, and undiagnosed. It is estimated that 42 % of patients
Bahrain leading the way (Fig. 4.4). The high rates with diabetes in Egypt have retinopathy, 5 % are
of diabetes are due in large part to the economic legally blind, and 22 % have peripheral neuropa-
growth of the last few decades and the subse- thy. Diabetes is also the leading cause of end-stage
quent change in diet and lifestyle of these popula- renal disease and leg amputation in Egypt [57]. It
tions. Saudi Arabia has a diabetes prevalence of is especially alarming that the prevalence of diabe-
23.9 % among adults. The prevalence is higher tes in Egypt has increased rapidly within a rela-
among Saudis residing in urban areas (25.5 %) tively short period of time from approximately 4.4
when compared to rural areas (19.5 %) [3]. There million in 2007 to 7.5 million in 2013. This number
is little gender difference with respect to diabetes is projected to rise to 13.1 million by 2035 (Fig. 4.5).
prevalence in the region. As expected, type 2 dia-
betes (T2D) cases predominate and constitute
approximately 9095 % of all patients with dia- Prevalence of Diabetes Complications
betes. However, it is worth mentioning that Saudi
Arabia has one of the worlds highest annual inci- Among patients with diabetes, the prevalence of
dence rates of type 1 diabetes in children, with microvascular and macrovascular complications is
31.4 new cases per 100,000 individuals per year relatively high in Arab populations. A recent cross-
[4]. This is the highest incidence of type 1 diabe- sectional study from Saudi Arabia showed that the
tes of any country outside of Europe. prevalence of peripheral neuropathy among patients
The countries with the largest number of people with diabetes was 19.9 % [8]. Among Saudi patients
affected by diabetes are Egypt (7.5 million), with T2D for at least 10 years, the prevalence of reti-
Pakistan (7 million), and Iran (4.5 million) (Fig. 4.4). nopathy was 31 % [9]. In Jordan, data from a national
These figures reflect the large populations of diabetes center showed that 45 % of patients had reti-
these countries. In Egypt, the prevalence of dia- nopathy, 33 % had nephropathy, and 5 % had a his-
betes is around 15.6 % among adults between 20 tory of amputation [10]. Diabetes is also frequently
and 79 years of age. The International Diabetes associated with psychological distress. A study from
Federation (IDF) estimates that 7.5 million UAE demonstrated that up to 33.8 % of patients with
diabetes had depression or anxiety or both [11].
Top 5 countries for number of pepole with diabetes Diabetic patients with depression are less likely to
(20-79 years), 2014 adhere to medical treatment and engage in self-care.
Countries/territories Millions Overall, the high prevalence of diabetic complica-
Egypt 7.593
tions in this region highlights the need for early
Pakistan 6.944
Iran (Islamic Republic of) 4.582
detection and prompt treatment of this disease.
Saudi Arabia 3.806
Sudan 3.007
Risk Factors for Type 2 Diabetes
Top 5 countries for diabetes comparative
prevalence (%) (20-79 years), 2014 in the MENA Region
Countries/territories %
Saudi Arabia 23.9 Obesity
Kuwait 23.1*
Bahrain 21.9* Obesity and physical inactivity are the two major
Qatar 19.8 risk factors for diabetes in the MENA region. The
United Arab Emirates 19
countries with the highest prevalence of obesity
Fig. 4.4 Diabetes in the MENA region, 2014 (Source: are Saudi Arabia, United Arab Emirates, and
IDF Atlas, sixth edition) Egypt (Fig. 4.6). Over the last few decades, there
4 Type 2 Diabetes in the Middle East and North Africa (MENA) 53

Fig. 4.5 The rising a Prevalence of T2D in Egypt

prevalence of type 2 20%
diabetes in Egypt. (a) 18%
Projected trends of the
prevalence rate of T2D in
Egypt. (b) The projected 14%
trend of the total number 12%
of patients with T2D in 10%
Egypt 8%
2007 2011 2030
Number of T2D patients in Egypt
(in millions)



2007 2013 2035

has been a dramatic increase in obesity, particu- among children has been reported in Bahrain
larly among Arab women. Results from a national (38.5 %), while the lowest was reported in Iran
survey conducted in Saudi Arabia showed the (3 %) [17]. Cross-sectional data from Saudi
prevalence of obesity was 44 % among women Arabia showed that a very high proportion of
and 26 % among men [12]. Similar results were Saudi adolescents (84 % of males and 91.2 % of
seen in the 2008 Egypt Demographic and Health females) spent more than 2 h on TV or video
Survey, which assessed the nutritional status of screen daily and almost half of the males and
adults aged 1559 years old, and found that three-quarters of the females did not meet daily
approximately 50 % of men and 6580 % of physical activity guidelines [18]. In Egypt, data
women were overweight or obese [13]. The from the most recent Demographic and Health
Egyptian National Hypertension Survey pro- Survey show that 35 % of males and 36 % of
gram, which was conducted in six Egyptian gov- females aged 519 are overweight [13]. These
ernorates and included 2,313 adults older than trends are largely the result of an unhealthy eat-
25 years of age, showed that 50 % of surveyed ing pattern and decreased physical activity that
individuals had central obesity. This was shown has emerged over the last few decades.
to be strongly associated with increased risk of
diabetes and cardiovascular disease [1416].
An alarming trend is the increased rates of Genetics of Diabetes
obesity among children and adolescents. Obesity
during childhood is a risk factor for obesity and Central adiposity is common among Arab popula-
related chronic diseases during adulthood, such tions and genetic factors are likely to play a role in
as cardiovascular disease, diabetes mellitus, and the development of obesity and diabetes. The genet-
hypertension. The highest prevalence of obesity ics underlying T2D is multifactorial and complex in
54 Y. Khan and O. Hamdy

30.3 28.8
30 28.9


15 13.5

















Fig. 4.6 Prevalence (%) of obesity in Arab countries (Source: 2010 WHO statistics)

nature. Although genetic studies from the region are computers and television have all contributed to a
limited, there is evidence that variations in the sedentary lifestyle. Low levels of physical activity
ADIPOQ gene, which encodes adiponectin, are are prevalent throughout the region. In a cross-sec-
associated with increased BMI and waist circumfer- tional study of adults in Saudi Arabia aged 3070,
ence [19]. Adiponectin is released by adipocytes the prevalence of inactivity was very high at 96.1 %
and is found in lower levels among obese subjects. [24]. In Cairo, absence of weekly physical activity
Recent studies from Jordan and Tunisia have also was reported among 81 % of the 4,918 households
shown that variants in the ADIPOQ gene strongly surveyed in 1995 [13]. The majority of individuals
correlate with risk of T2D [20, 21]. In the Lebanese did not engage in any physical activity or walk
population, mutations in both CDKAL1 and regularly. Physical activity levels were especially
IGF2BP2 genes have been shown to be associated low among women [24]. The major reasons for
with T2D [22, 23]. Interestingly, variants of the decreased physical activity, particularly among
TCF7L2 gene, which have been associated with Egyptian women, include the general lack of exer-
T2D across multiple ethnic groups, have only a cise facilities, overcrowded urban cities, and poor
weak association with diabetes in Arab populations. physical education in schools [13]. Furthermore,
Another consideration is that the genetic risk is urban planning in many cities does not encourage
increased among populations that have traditions of or support an active lifestyle. The lack of regular
consanguineous marriages. Certainly more research physical activity represents a major public health
is needed to further our understanding of the genetic concern that needs to be addressed by each indi-
basis of diabetes in this region. vidual country in a culturally appropriate way.

Physical Inactivity Chronic Hepatitis C Infection

Physical inactivity is the other major risk factor for The burden of hepatitis C virus (HCV) is growing
the development of T2D in MENA. Rapid eco- in the MENA region. If untreated, HCV can
nomic development, increased availability of cars, lead to liver cirrhosis, hepatocellular cancer, and
access to migrant labor, and widespread use of death. Egypt has the highest prevalence of chronic
4 Type 2 Diabetes in the Middle East and North Africa (MENA) 55

hepatitis C infection in the world. This is attributed can affect multiple pathways involved in glucose
to the mass intravenous treatment campaign of bil- regulation. Organophosphorus and organochlorine
harziasis between 1960 and 1980 with the use of pesticides in particular lead to a deleterious effect
poorly sterilized needles. The Egypt Demographic on glucose metabolism and insulin secretion. It is
and Health Survey showed that 15 % of Egyptians hypothesized that the involved mechanisms include
are serologically positive for HCV antibodies and oxidative stress, pancreatitis, and inhibition of cho-
10 % have active infection [13]. Meanwhile in the linesterase [30, 31].
Gulf region, infection rates are relatively low. In Pesticides are widely used in the MENA
Qatar, 1.1 % of the population carries the virus. region, as 43 % of the population lives in rural
The prevalence of T2D among patients with areas and depends on farming for livelihood.
HCV is 1333 %. A meta-analysis of the associa- Pesticide usage is high in Lebanon, Kuwait, and
tion of HCV and T2D showed that patients with Egypt. Egypt ranks fifth among African nations
HCV are more likely to develop T2D. The odds with respect to pesticide consumption. The most
ratio was particularly high among male patients commonly used pesticides in Egypt are dichloro-
(OR, 1.26; 95 % CI, 1.031.54) and those older diphenyltrichloroethane (DDT), which is an
than 40 years of age (OR, 7.39; 95 % CI, 5.82 organochlorine compound, and chlorpyrifos and
9.38). Not only does HCV increase the risk of malathion, which are organophosphorus com-
diabetes, it is also associated with poor glycemic pounds [3235]. Growing evidence suggests a
control and increased prevalence of diabetes strong association between exposure to these
complications. A study of 438 patients with T2D pesticides and increased risk for insulin resis-
(113 Egyptians and 325 Kuwaitis) showed that tance and T2D. Raafat et al. found a positive
poor glycemic control was mostly seen in patients association between chronic exposure to mala-
positive for HCV. In another cross-sectional thion and insulin resistance [36]. Another study
study of 489 patients with T2D who attended an showed a U-shaped dose-response association
outpatient clinic and dialysis unit in Egypt, the between features of metabolic syndrome, such as
prevalence of HCV infection was 12.9 % among high body mass index, dyslipidemia (increased
patients attending the outpatient clinic and 18.7 % serum triglycerides, increased LDL cholesterol,
among patients on dialysis [2528]. decreased HDL cholesterol), and insulin resis-
Early treatment of HCV can help delay or pre- tance with exposure to various pesticides [37].
vent T2D. It has been shown that effective elimi- Currently, pesticide regulations are inadequate
nation of HCV in patients with prediabetes and a greater commitment to food safety is
improved glycemic tolerance and significantly needed to protect the health of consumers.
reduced A1C levels. Prospectively, 34.8 % of
patients who were treated became normoglyce-
mic and only 5.5 % developed diabetes. The sus- Cultural Factors Affecting Type 2
tained response to antiviral therapy was the only Diabetes in the MENA Region
independent predictor of improved glycemic
control [29]. With the development of newer and Dietary Pattern
effective medications for HCV, the risk for T2D
may significantly reduce in Egypt. Over the last three decades, the MENA region has
experienced a major shift in its traditional eating
patterns, referred to as the nutrition transition,
Pesticide Exposure due to urbanization and rapid economic develop-
ment. The traditional diet was rich in vegetables,
There is growing evidence that pesticide use is asso- legumes, fruits, whole wheat bread, and fish, with
ciated with T2D. Exposure to pesticides may occur low to moderate in amount of animal protein.
either directly among farmers and applicators or Presently, a dietary pattern with high consump-
indirectly due to chronic exposure to low levels of tion of saturated fat, refined carbohydrates,
pesticides through contaminated food. Pesticides processed meat, and sugar-sweetened beverages
56 Y. Khan and O. Hamdy

has emerged throughout the region. This pattern countries, reduced exposure to sunlight with tra-
has been associated with an increased risk of ditional clothing among women contributes to
T2D. Meanwhile, diets characterized by high vitamin D deficiency, which has been linked to
intakes of vegetables, fruits, and whole grains have increased rates of obesity and T2D [40].
been shown to be protective against T2D [38]. Women are facing more obstacles to physical
Recent meta-analyses of observational and activity compared to men. In Bahrain, for exam-
prospective cohort studies have shown a strong ple, the main barriers to exercise perceived by
association between high consumption of white women were home commitments (49 %), care of
rice and an increased risk of T2D [39]. High con- children (36 %), and negative attitudes by family
sumption of trans fat has also been shown to be a members toward women practicing exercise/
risk factor for cardiovascular disease. Egypt and sports (24 %) [41, 42]. Of the women studied, the
Pakistan are the worlds highest consumers of majority believed that there is gender discrimina-
this unhealthy type of fat. Partially hydrogenated tion when it comes to sports, since sports and
oil, which contains trans fat, is also used fre- other recreational facilities are provided mostly
quently for daily cooking and for preparing fried for men [42]. Furthermore, in the Middle East,
foods throughout the region. women athletes continue to struggle for womens
Sociocultural factors also play an important role participation in sports at the national and interna-
in the pattern of food consumption. With longer tional level. In 2012, Sarah Attar and Wojdan
working hours and more women entering the work- Shaherkani made history when they became the
force, there has been a gradual shift toward larger first women to compete for Saudi Arabia in the
and later dinner. Dining out is also on significant London Olympics.
rise, especially among youth. Western fast-food
restaurants have also proliferated throughout the
MENA region. Together with a strong culture of Smoking
hospitality and frequent family gatherings around
food, these trends have nudged society toward con- Smoking represents a major public health crisis
sumption of bigger portions and no doubt are con- in the MENA region. Smoking is directly linked
tributing to the increased prevalence of obesity and to increased incidence of microvascular and mac-
T2D in the MENA region. rovascular diseases in patients with diabetes. In
Jordan, 43.4 % of men smoke, which represents
the highest prevalence in the region. Among
Sedentary Lifestyle women, Lebanon leads the way with a smoking
prevalence of 21.2 %. Egypt represents one of the
Low levels of physical activity are further exac- largest tobacco markets in the world, with over
erbating the problem of obesity and diabetes. 20 billion cigarettes smoked annually and a
The increased number of cars, vast expansion smoking prevalence of 39.7 % among adult men.
of cities, and hot climate contribute to decreased Smoking is still rare among Egyptian women.
physical daily activity. In addition, there are Despite high taxes on cigarettes and increasing
various cultural and social norms that may dis- public health education, smoking remains a com-
courage exercise. In Egypt, there is a tendency mon unhealthy habit in the MENA region [43].
to avoid exercise in public areas, while few are
able to afford membership in private athletic
facilities. Overall, exercise facilities are limited Health Illiteracy
and community sports facilities are scarce. Even
in schools, reduced time for physical activity Health literacy may be defined as the degree to
and limited availability of spaces for sports rein- which an individual has the capacity to obtain and
forces a sedentary lifestyle for Egyptian youth. It understand basic health information to make
is also worth mentioning that in Egypt and Gulf appropriate health decisions. Health illiteracy is
4 Type 2 Diabetes in the Middle East and North Africa (MENA) 57

common in the MENA region. Obesity is fre- effective prevention. Unfortunately, many
quently regarded as a cosmetic problem and rarely patients consider their disease and its complica-
viewed as a disease. Most patients with T2D tions as inevitable and feel powerless to alter the
believe that diabetes should only be treated by course of the disease. A recent study from Egypt
oral medications and often resist insulin injections showed that improvement with medication adher-
when indicated. Routine daily glucose monitoring ence can be achieved through improving patient
is essentially nonexistent or limited due to cost or education about the disease, simplifying the drug
fear of frequent finger sticks. Furthermore, it has regimen, and reducing medication cost [46].
been demonstrated that patients rarely change
their diet or exercise habits following a diagnosis
of diabetes. A recent survey of 575 patients from Healthcare Quality
a diabetes outpatient clinic in the UAE showed
that the majority (72 %) of patients had negative The high prevalence of diabetes and other chronic
attitudes toward having the disease, and 31 % of diseases has placed a great burden on the regions
patients had poor knowledge of diabetes [44]. health systems. Many patients with diabetes are
In a study from Egypt, the majority of 560 either treated in the limited number of public hos-
patients with T2D surveyed believed that T2D is pitals, in the private healthcare sector through out-
an infectious disease originally caused by stress. of-pocket fee for service, or in the scarce diabetes
Moreover, only 38.4 % of patients had a positive centers in major cities. Furthermore, patients with
attitude toward self-management. Interestingly, diabetes face many barriers to care including cost,
these study patients believed that the efficacy of access to physicians and diabetes educators, and
herbal medicine in treating T2D was sufficiently access to medicines such as insulin.
high and neglected the positive role of regular Human resources for diabetes care are often
exercise in diabetes management. Most of inadequate. Overall, there is a shortage of physi-
patients also believed that patients suffering from cians, nurses, and health educators in the region.
polyuria should reduce their volume of drinking For example, a study in Saudi Arabia found that
water [45]. All of these factors result in poor gly- less than 10 % of health centers were staffed
cemic control, late diagnosis, and increased prev- with diabetes educators, and only 40 % of
alence of diabetes complications. patients with retinopathy were referred to eye
clinics [47]. Another study examined the deliv-
ery of health education to T2D patients in four
Poor Adherence healthcare centers (two rural and two urban) in
Alexandria, Egypt. Although diabetes knowl-
In the MENA region, as elsewhere, nonadher- edge was satisfactory among the 88 physicians
ence with diabetes management is common and surveyed in this study, 95.5 % of physicians
is one of the main factors for poor glycemic con- from rural areas and 89.8 % of those from urban
trol. Daily testing of blood glucose is very rare health centers neglected the fundamental role of
and frequent omission of insulin injections is patient education and regular exercise in man-
common, even in patients with type 1 diabetes. aging T2D [45]. All of these factors contribute
Even among educated patients, poor adherence to suboptimal quality of care for patients with
with healthy eating and physical activity is com- diabetes.
mon. Furthermore, fear of hypoglycemia among
patients treated with insulin leads to suboptimal
control and frequent reduction in insulin dosing Prevention of T2D
either intentionally or upon recommendation of
treating physicians. Frequently, diagnosis of dia- Type 2 diabetes is a largely preventable disease.
betes complications, especially retinopathy and Research has shown that lifestyle modification
nephropathy, is too late and does not allow for can significantly reduce the incidence of type 2
58 Y. Khan and O. Hamdy

diabetes by up to 58 % [48]. Early diagnosis and further insight into the prevention and treatment
prevention of diabetes is a key priority for coun- of this disease.
tries of the MENA region. The Arab Taskforce To curb the diabetes epidemic, public health
for Obesity was created in 2010 to limit the rise strategies that focus on primary prevention
of obesity among Arab countries. This repre- through promotion of a healthy diet and lifestyle
sented a collaboration of 14 Arab countries. The should be prioritized. Primary care is now seen as
goals of the taskforce are to promote healthy eat- having a key role in prevention and early detec-
ing, increase physical activity, raise awareness tion of diabetes. Traditionally, diabetes has been
about obesity and its complications, and develop managed by specialist centers, but the majority of
national guidelines for controlling the rise in obe- people with diabetes can be effectively managed
sity, especially in children. by multidisciplinary teams of physicians, diabe-
Effective screening is one of the most impor- tes educators, nurses, and pharmacists. Countries
tant factors in diabetes prevention and control. should develop and organize their health systems
The Weqaya program in Abu Dhabi is an exam- to provide integrated and continuous care for
ple of a successful screening program. The pro- patients from prediabetes to severe diabetes with
gram was launched in 2008, and to date over complications.
94 % of the national population has been screened
for cardiovascular risk factors, including diabe-
tes, using A1c criteria [49]. Approximately 45 % Strategies to Improve Diabetes Care
of the screened population was found to have dia- in the MENA Region
betes or prediabetes. In Oman, a diabetes risk
score was developed that used age, waist circum- As outlined in this chapter, the diabetes epidemic
ference, body mass index, family history of dia- is driven by a complex multitude of factors. As
betes, and hypertension status to correctly awareness of diabetes in the region rises, a tar-
identify individuals at high risk of type 2 diabetes geted multi-sector response is necessary to
in a community setting. Interestingly, testing address this global crisis. Fundamental changes in
with Dutch, Thai, and Finnish diabetes risk scores public policies and health systems are required.
showed poor performance in this population [50]. The following points should be urgently
This suggests that customized screening tools considered:
may be of benefit in Arab populations. Ideally,
screening for diabetes should be part of a larger 1. Screen for diabetes.
national prevention program. Abu Dhabi, Qatar, Health authorities should implement efficient
UAE, and Kuwait are examples of countries that screening programs for high-risk individuals,
are leading the way with national strategies for especially adults who are overweight and
diabetes prevention. obese with a positive family history of diabe-
tes. Focusing on primary care and preventive
services will avoid costly complications of
Future Directions diabetes down the road.
2. Create a National Diabetes Prevention
Although the prevalence of diabetes is increasing Program.
throughout the MENA region, data on obesity Develop a comprehensive and accessible life-
and diabetes in many countries are lacking. style program with simple methods to facili-
Healthcare research is not seen as a priority in tate implementation on large scale. Increase
many countries of the region. Gaps exist in our public awareness of diabetes and emphasize
knowledge of epidemiological variation of diabe- that early diagnosis can reduce the risk of dia-
tes, as Arab populations are widely distributed betes and its complications. Addressing
across both Asia and Africa. Research on social childhood obesity and obesity among women
and cultural barriers to diabetes care can add should be a priority of such programs.
4 Type 2 Diabetes in the Middle East and North Africa (MENA) 59

3. Improve diabetes education. 7. Develop clinical guidelines.

Diabetes education remains the cornerstone of Guidelines should be published for diabetes
diabetes management. General diabetes and management that are culturally suitable for
nutrition education should be conducted at patients. Such guidelines should consider the
schools, community centers, and healthcare unique risk factors highlighted in this chapter.
facilities. Public and private media can also play Guidelines can help standardize clinical prac-
vital role in disseminating accurate informa- tice to improve patient care. There is a great
tion about diabetes to the public. The goal is to opportunity for local diabetes associations
encourage citizens to be engaged in a healthy and scientific bodies to champion the devel-
lifestyle and also to reduce consumption of pro- opment of these guidelines and audit their
cessed food, refined carbohydrates, and trans fat. implementation.
4. Strengthen clinical care for diabetes.
Improve human resources for healthcare. The
concept of a diabetes management team should Summary and Conclusions
be introduced that includes a certified diabetes
educator (CDE) and registered dietitian (RD). Diabetes is a growing public health problem in the
Health authorities should develop or encour- MENA region. Its high prevalence continues to
age training programs that aim to prepare rise due to the increased prevalence of central obe-
enough CDEs and RDs to treat the growing sity, sedentary lifestyle, change in diet patterns,
number of patients with diabetes. Education increased prevalence of hepatitis C, and possibly
programs should emphasize self-management the increased use of uncontrolled pesticides.
and dispel fears about finger sticks and insulin Smoking among men, health illiteracy, and poor
injections. Physicians and patients alike should adherence increase the frequency of diabetes com-
be encouraged to be involved in this process. plications. Health authorities, through a limited
In addition, mobile technology in the form of healthcare budget, are striving to improve diabetes
apps, reminders, and education material can care, but many strategies and guidelines for stan-
also be used to enhance a comprehensive dia- dard of care are still needed to augment this effort.
betes management plan.
5. Treat hepatitis C infection.
The risk of hepatitis C infection should be References
reduced to minimize the health and economic
burden of diabetes. Early management with a 1. International Diabetes Federation. IDF diabetes. 7th
focus toward elimination of HCV in infected ed. Brussels: International Diabetes Federation; 2015.
patients can also reduce T2D risk. Newer and 2. NCD Risk Factor Collaboration. Worldwide trends in
effective medications should be available at diabetes since 1980: a pooled analysis of 751
reduced price or provided through the national population-based studies with 4.4 million partici-
insurance service. pants. Lancet. 2016;387(10027):151330.
3. Al-Nozha MM, Al-Maatouq MA, Al-Mazrou YY,
6. Regulate pesticide use. Al-Harthi SS. Diabetes mellitus in saudi arabia. Saudi
Implementation of an effective pesticide regula- Med J. 2004;25:160310.
tory program, with regular surveying of pesti- 4. Habeb AM, Al-Magamsi MS, Halabi S, Eid IM,
cide residues in drinking water and food, should Shalaby S, Bakoush O. High incidence of childhood
type 1 diabetes in Al-Madinah, North West Saudi
be mandated. Health education is also war- Arabia (20042009). Pediatr Diabetes. 2011;12(8):
ranted to reduce uncontrolled exposure to pesti- 67681.
cides by minimizing the misuse and improper 5. Al-rubeaan K. International journal of diabetes mel-
handling of pesticides and providing protective litus type 2 diabetes mellitus red zone. Int J Diabetes
Mellitus. 2010;2:12.
personal equipment. Special attention should be 6. Jain S, Saraf S. Type 2 diabetes mellitusits global
made to avoid the involvement of women and prevalence and therapeutic strategies. Diabetes Metab
children in pesticide application. Syndr: Clin Res Rev. 2010;4:4856.
60 Y. Khan and O. Hamdy

7. Whiting DR, Guariguata L, Weil C, et al. IDF diabetes haplotypes in the adiponectin gene contribute to the
atlas: global estimates of the prevalence of diabetes for genetic risk for type 2 diabetes in Tunisian Arabs.
2011 and 2030. Diabetes Res Clin Pract. 2011;94: Diabetes Res Clin Pract. 2012;97(2):2907.
31121. 22. Nemr R, Almawi AW, Echtay A, Sater MS, Daher HS,
8. Wang DD, Bakhotmah BA, Hu FB, Alzahrani Almawi WY. Replication study of common variants
HA. Prevalence and correlates of diabetic peripheral in CDKAL1 and CDKN2A/2B genes associated with
neuropathy in a saudi arabic population: a cross- type 2 diabetes in Lebanese Arab population. Diabetes
sectional study. PLoS One. 2014;9(9):e106935. Res Clin Pract. 2012;95(2):e3740.
9. El-Asrar AMA, Al-Rubeaan KA, Al-Amro SA, 23. Nemr R, Echtay A, Dashti EA, Almawi AW,
Kangave D, Moharram OA. Risk factors for diabetic Al-Busaidi AS, Keleshian SH, Irani-Hakime N,
retinopathy among Saudi diabetics. Int Ophthalmol. Almawi WY. Strong association of common variants
1998;22(3):15561. in the IGF2BP2 gene with type 2 diabetes in Lebanese
10. Jbour AS, Jarrah NS, Radaideh AM, Shegem NS, Arabs. Diabetes Res Clin Pract. 2012;96(2):2259.
Bader IM, Batieha AM, Ajlouni KM. Prevalence and 24. Al-Nozha MM, Al-Hazzaa HM, Arafah MR,
predictors of diabetic foot syndrome in type 2 diabe- Al-Khadra A, Al-Mazrou YY, Al-Maatouq MA, Khan
tes mellitus in Jordan. Saudi Med J. 2003;24(7): NB, Al-Marzouki K, Al-Harthi SS, Abdullah M,
7614. Al-Shahid MS. Prevalence of physical activity and
11. El-Rufaie OE, Bener A, Ali TA, Abuzeid inactivity among Saudis aged 3070 years. A
M. Psychiatric morbidity among type II diabetic population-based cross-sectional study. Saudi Med
patients: a controlled primary care survey. Prim Care J. 2007;28(4):55968.
Psychiatry. 1997;3:18994. 25. Miller FD, Abu-Raddad LJ. Evidence of intense
12. AlQuaiz AM, Siddiqui AR, Qureshi RH, Fouda MA, ongoing endemic transmission of hepatitis C virus
AlMuneef MA, Habib FA, Turkistani IM. Women in Egypt. Proc Natl Acad Sci U S A. 2010;107:
health in Saudi Arabia: a review of non-communicable 1475762.
diseases and their risk factors. Pak J Med Sci. 26. Chehadeh W, Abdella N, Ben-Nakhi A, et al. Risk
2014;30(2):422. factors for the development of diabetes mellitus in
13. EL-Zanaty F, Way A. Egypt demographic and health chronic hepatitis C virus genotype 4 infection.
survey 2008. Cairo: Ministry of Health, El-Zanaty and J Gastroenterol Hepatol. 2009;24:428.
Associates, and Macro International; 2009. p. 431. 27. Chehadeh W, Kurien SS, Abdella N, et al. Hepatitis C
14. Population Council. virus infection in a population with high incidence of
pdfs/2010PGY_SYPEFinalReport_FrontMatter. type 2 diabetes: impact on diabetes complications.
pdf. J Infect Public Health. 2011;4:2006.
15. Ibrahim MM, Elamragy AA, Girgis H, et al. Cut off 28. Greca LF, Pinto LC, Rados DR, et al. Clinical features
values of waist circumference & associated cardiovas- of patients with type 2 diabetes mellitus and hepatitis
cular risk in Egyptians. BMC Cardiovasc Disord. C infection. Braz J Med Biol Res Rev Bras Pesqui
2011;11:53. Md Biol Soc Bras Biofs [et al]. 2012;45:28490.
16. Herman WH, Ali MA, Aubert RE, et al. Diabetes mel- 29. Huang JF, Yu ML, Huang CF, et al. The outcomes of
litus in Egypt: risk factors and prevalence. Diabet glucose abnormalities in pre-diabetic chronic hepati-
Med. 1995;12(12):112631. tis C patients receiving peginterferon plus ribavirin
17. Mirmiran P, Sherafat-Kazemzadeh R, Jalali-Farahani therapy. Liver Int. 2012;32:9629.
S, Azizi F. Childhood obesity in the Middle East: a 30. FAO/RNEA. Report of the workshop on implementa-
review/Revue sur lobsit de lenfant au Moyen- tion of the International Code of Conduct on the dis-
Orient. East Mediterr Health J. 2010;16(9):1009. tribution and use of pesticides in the Near East, 1621
18. Al-Hazzaa HM, Abahussain NA, Al-Sobayel HI, Sept 1995, Cairo; 1995.
Qahwaji DM, Musaiger AO. Physical activity, seden- 31. WRI. Pesticides and the immune system: the public
tary behaviors and dietary habits among Saudi adoles- health risks. Washington, DC: World Resources
cents relative to age, gender and region. Int J Behav Institute; 1996.
Nutr Phys Act. 2011;8(1):140. 32. Ambio O. Pesticide use, habits and health awareness
19. Zadjali F, Al-Yahyaee S, Hassan MO, Albarwani S, among Egyptian farmers. Ann Occup Hyg. 1996;
Bayoumi RA. Association of adiponectin promoter 40(5):499509.
variants with traits and clusters of metabolic syndrome 33. Wessels D, Barr DB, Mendola P. Use of biomarkers to
in Arabs: family-based study. Gene. 2013;527(2): indicate exposure of children to organophosphate pes-
6639. ticides: implications for a longitudinal study of chil-
20. Alkhateeb A, Al-Azzam S, Zyadine R, Abuarqoub drens environmental health. Environ Health Perspect.
D. Genetic association of adiponectin with type 2 dia- 2003;111(16):193946.
betes in Jordanian Arab population. Gene. 2013; 34. Mansour SA. Pesticide exposureEgyptian scene.
512(1):613. Toxicology. 2004;198(13):91115.
21. Mtiraoui N, Ezzidi I, Turki A, Chaieb A, Mahjoub T, 35. El-Sebae AH, Zeid MA, Saleh MA. Status and envi-
Almawi WY. Single-nucleotide polymorphisms and ronmental impact of toxaphene in the third worlda
4 Type 2 Diabetes in the Middle East and North Africa (MENA) 61

case study of African agriculture. Chemosphere. M, Rashid N, El Hasnaoui A, BREATHE Study

1993;27(10):206372. Group. Smoking habits in the Middle East and North
36. Raafat N, Abass MA, Salem HM. Malathion expo- Africa: results of the BREATHE study. Respir Med.
sure and insulin resistance among a group of farmers 2012;106:S1624.
in Al-Sharkia governorate. Clin Biochem. 2012; 44. Al-Maskari F, El-Sadig M, Al-Kaabi JM, Afandi B,
45:15915. Nagelkerke N, Yeatts KB. Knowledge, attitude and
37. Lee DH, Steffes MW, Sjdin A, et al. Low dose practices of diabetic patients in the United Arab
organochlorine pesticides and polychlorinated biphe- Emirates. PLoS One. 2013;8(1):e52857.
nyls predict obesity, dyslipidemia, and insulin resis- 45. Koura MR, Khairy AE, Abdel-Aal NM, et al. The role
tance among people free of diabetes. PLoS One. of primary health care in patient education for diabe-
2011;6(1), e15977. tes control. J Egypt Public Health Assoc. 2001;
38. Brunner EJ, Mosdl A, Witte DR, Martikainen P, 76(0013-2446 (Print)):24164.
Stafford M, Shipley MJ, Marmot MG. Dietary patterns 46. Shams ME, Barakat EA. Measuring the rate of thera-
and 15-y risks of major coronary events, diabetes, and peutic adherence among outpatients with T2DM in
mortality. Am J Clin Nutr. 2008;87(5):141421. Egypt. Saudi Pharm J. 2010;18(4):22532.
39. Hu EA, Pan A, Malik V, Sun Q. White rice consump- 47. Al-Ahmadi H, Roland M. Quality of primary health
tion and risk of type 2 diabetes: meta-analysis and care in Saudi Arabia: a comprehensive review.
systematic review. BMJ. 2012;344:e1454. International J Qual Health Care. 2005;17(4):331.
40. Fields J, Trivedi NJ, Horton E, et al. Vitamin D in the 48. Diabetes Prevention Program Research Group.
Persian Gulf: integrative physiology and socioeco- Reduction in the incidence of type 2 diabetes with
nomic factors. Curr Osteoporos Rep. 2011;9:24350. lifestyle intervention or metformin. N Engl J Med.
41. Musaiger AO. Overweight and obesity in the Arab coun- 2002;346(6):393.
tries: the need for action. Arab Cent Nutr. 2007;1:18. 49. Hajat C, Harrison O, Al Siksek Z. Diagnostic testing
42. Musaiger AO, Al-Ansari MS. Barriers to practicing for diabetes using HbA1c in the Abu Dhabi popula-
physical activity in the Arab countries, in nutrition tion Weqaya: the Abu Dhabi cardiovascular screening
and physical activity in the arab countries of the near program. Diabetes Care. 2011;34(11):24002.
east. Musaiger AO and Meladi SS, (eds). FAO/Cairo 50. Al-Lawati JA, Tuomilehto J. Diabetes risk score in
Regional Office, Cairo, Egypt. 2000. Oman: a tool to identify prevalent type 2 diabetes
43. Khattab A, Javaid A, Iraqi G, Alzaabi A, Kheder AB, among Arabs of the Middle East. Diabetes Res Clin
Koniski ML, Shahrour N, Taright S, Idrees M, Polatli Pract. 2007;77(3):43844.
Diabetes in China and the Western
Pacic Region 5
Juliana C.N. Chan, Elaine Y.K. Chow,
and Andrea O.Y. Luk

Introduction lion people, have diabetes which is expected to

increase to 201.8 million in the next 20 years.
According to the International Diabetes While China has the biggest diabetic population
Federation (IDF) Atlas, 382 million people have with 98 million people affected, the WP Region
diabetes in 2013, and by 2035, this will rise to also has areas with the highest prevalence, such
582 million, with China, India and the United as some Pacific Islands with a diabetes preva-
States of America (USA) as the top three coun- lence of 30 %. A major concern is the high preva-
tries contributing to half of the global population lence of undiagnosed diabetes and prediabetes
of diabetes. The Western Pacific (WP) Region is which can lead to late presentation with expen-
the worlds most populous region with 39 coun- sive and difficult-to-treat complications.
tries and territories and home to 35 % of the Figure 5.1 summarises the IDF estimates regard-
world population. It has the largest country, ing the prevalence, disease burden and distribu-
China, with one billion people and the smallest tion of type 1 and type 2 diabetes in different age
Pacific Island nations with less than 1000 people. and gender groups as well as countries within the
It is also a region of transition with enormous WP Region in 2013 (Table 5.1) [1].
diversity in terms of ethnicity, demographics, Adding to this health-care challenge are the
cultures, politics, religions, technological and rapid societal transition and changing demograph-
socioeconomic development as well as health- ics occurring within an unprepared health-care
care systems, all of which are implicated in the system designed for provision for acute and epi-
causality and consequences of diabetes [1]. sodic rather than chronic care. In many emerging
According to the latest estimates by the IDF, economies, the lack of societal, environmental
8.6 % of adults in the WP Region, i.e. 138.2 mil- and financial protection has led to the widening
social disparity which has exposed the genetically
vulnerable and socially disadvantaged to develop
this silent killer often due to lack of information,
awareness and intervention [2, 3]. In this chapter,
J.C.N. Chan, MD, FRCP (*) E.Y.K. Chow, MB, we will review the epidemiology, aetiology, mor-
ChB, PhD A.O.Y. Luk, MBBS, MRCP bidity, diagnosis, treatment and care delivery for
Department of Medicine and Therapeutics, Hong
Kong Institute of Diabetes and Obesity and Li Ka diabetes focusing in China and WP Region. We
Shing Institute of Health Sciences, The Chinese will also discuss the current unmet needs and pos-
University of Hong Kong, Prince of Wales Hospital, sible solutions for the prevention and control of
Shatin, Hong Kong, SAR, China diabetes in this populous region.

Springer International Publishing Switzerland 2017 63

S. Dagogo-Jack (ed.), Diabetes Mellitus in Developing Countries and Underserved Communities,
DOI 10.1007/978-3-319-41559-8_5
64 J.C.N. Chan et al.

Diabetes 2010 60,000

Diabetes prevalence (%)

12 2007 50,000

GDP (million RMB)

2000 40151
8 5.5 30,000
26581 20,000
4 2.5
2 0.9 10,000
0 0
1978 1983 1988 1993 1998 2003 2008 2013

Fig. 5.1 The growth of gross domestic product (GDP) and prevalence of diabetes increased in a parallel manner in
China during the last three decades [3]

Rapid Transition and the Obesity decades, while the GDP has increased by 100-
Epidemic fold, the prevalence of diabetes has also increased
by tenfold. In the 2010 National Survey of
The impact of rapid urbanisation on the diabetes 98,658 adults and using both 75 g oral glucose
epidemic is exemplified by the parallel increase tolerance test (OGTT) and glycated haemoglo-
in gross domestic product (GDP) and diabetes bin (A1c) as diagnostic criteria, one in nine
prevalence in China (Fig. 5.1). In less than three Chinese had diabetes, one in three had obesity
(general or central) closely associated with
hypertension and dyslipidemia and one in two
Table 5.1 Key facts and figures about diabetes in the had prediabetes. Amongst those with diabetes,
Western Pacific Region [1] only 30 % were diagnosed and amongst the diag-
At a glance 2013 2035 nosed, only 30 % were treated and amongst the
Total population (millions) 2278 2476 treated, only 40 % were controlled, defined as
Adult population (2079 years, 1613 1818 A1c less than 7 % [4].
millions) According to the Diabetes Epidemiology
Diabetes (2079 years)
Collaborative Analysis of Diagnostic Criteria in
Regional prevalence (%) 8.6 11.1
Asia (DECODA) data, 3050 % of subjects in the
Comparative prevalence (%)* 8.1 8.4
WP Region had metabolic syndrome, character-
Number of people with diabetes 138.2 201.8
ised by central obesity and clustering of cardio-
IGT (2079 years) metabolic risk factors, and a strong predictor for
Regional prevalence (%) 6.8 9.0 future development of diabetes [5]. In 2008,
Comparative prevalence (%)* 6.6 7.8 6070 % of people in Tonga, Nauru and Cook
Number of people with IGT (millions) 110.1 164.5 Islands were considered obese [6]. In 2004, the
Type 1 diabetes (014 years) sex-standardised diabetes prevalence was reported
Number of children with type 1 32.5 to be 13.0 % in men and 14.4 % in women in
diabetes (thousands) Nauru. The age-specific prevalence rose from
Number of newly diagnosed cases 5.3 4.5 % in the 1524 age group to 42.7 % in the
per year (thousands) 5564 age group [7]. Tables 5.2 and 5.3 sum-
Health expenditure due to diabetes (2079 years, USD) marise the prevalence and number of people with
Total health expenditure, R = 2#, 88.4 98.4 diabetes in the WP countries based on the IDF fig-
ures published in 2013 [8].
# Rapid quantitative and qualitative changes in
R=2 (the diabetes cost ratio assumes that healthcare costs
for people with diabetes are on average two fold higher nutritional intake, notably a shift from a
than people without diabetes). traditional high-fibre, low-fat diet to an energy-
5 Diabetes in China and the Western Pacific Region 65

Table 5.2 Top ten countries in the Western Pacific Region Table 5.3 Top ten countries in the Western Pacific
in terms of number of subjects affected by diabetes [8] Region in terms of prevalence of diabetes [8]
Number of subjects (in 1000s), Prevalence (%),
Country/territory 2013 Country/territory 2013
1. China 98,407.379 1. Tokelau 37.49
2. Indonesia 8554.165 2. Federated States of Micronesia 35.03
3. Japan 7203.776 3. Marshall Islands 34.89
4. Republic of Korea 3323.903 4. Kiribati 28.77
5. Vietnam 3299.113 5. Cook Islands 25.66
6. Philippines 3256.215 6. Vanuatu 23.97
7. Thailand 3150.670 7. Nauru 23.29
8. Myanmar 1988.846 8. French Polynesia 22.41
9. Malaysia 1913.236 9. New Caledonia 19.49
10. Taiwan 1721.062 10. Guam 19.48

dense diet with increased meat and fat consump- Besides, Asians had the lowest body mass index
tion as well as reduced physical activity due to (BMI) and AIRg and developed diabetes only
mechanisation and increased use of mobile vehi- with a small increase in BMI compared to the
cles, are important drivers for this global epi- Caucasians and Africans [13].
demic [9]. In the islands of Vanuatu, researchers Compared to their Europid counterparts, peo-
reported close associations of economic develop- ple of Asian ethnicity are more prone to develop
ment with increased consumption of animal pro- diabetes for the same level of BMI or waist cir-
teins and simple carbohydrates, as well as cumference due to their propensity to store
increased tobacco and alcohol consumption in excessive body or visceral fat [14]. Even amongst
men in areas with high tourism [10]. In China, lean subjects and given the same amount of vis-
during the last two decades, changes in food tech- ceral fat, Asians were more insulin resistant than
nology with production of high-calorie foods Europids due to increased concentration of free
together with reduced physical activity have con- fatty acids and inflammatory markers [15].
tributed to the rising burden of obesity in both Using insulin clamp studies, researchers from
adults and children [11]. Thailand reported insulin deficiency as the pre-
dominant feature in lean subjects and insulin
resistance and in the overweight and obese sub-
Unique Aspects jects [16].
of Pathophysiology: Beta-Cell In autopsy studies, Korean researchers have
Function Versus Visceral Obesity reported close correlations between BMI and
relative percentage of beta cells in both diabetic
Age, family history and obesity are the main risk (r = 0.55) and nondiabetic specimens (r = 0.35).
factors for diabetes [4] although inadequate beta- However, diabetic subjects had a lower percent-
cell response to overcome insulin resistance due age of beta cells than nondiabetic subjects for
to factors such as ageing, obesity and inflam- the same BMI [17]. In healthy subjects, both
mation remains the primary culprit [12]. In an Chinese and South Asian subjects with normal
analysis of 74 study cohorts comprising 3813 glucose tolerance exhibit higher glucose excur-
individuals (19 African, 31 Caucasian and 24 sion during an OGTT with reduced rate of glu-
East Asian cohorts), the authors examined the cose disposal than their Caucasian counterparts
hyperbolic relationship between insulin sensitiv- [18]. This inherently low beta-cell function
ity index (IS) and acute insulin response to glu- together with propensity to develop obesity
cose (AIRg) in healthy cohorts. Amongst these which will impose metabolic stress has provided
three subpopulations, Asians had a steep and a highly plausible explanation for the high risk
non-linear relationship with small changes in one of diabetes in Asian populations undergoing
variable having large effect sizes on the other. rapid transition (Fig. 5.2).
66 J.C.N. Chan et al.

D Plasma Glucose (mmol/L)

SE Asian (n= 10) Insulin Sensitivity (euglycemic clamp)
4.5 Caucasian (n= 20)

M value (mmol/min/m2)
2.0 0.50
1.0 0.00
0.5 Caucasian Chinese SE Asian
0.0 n = 15 n=9 n=7
0 15 30 45 60 75 90 105 120
Time (minutes)
25 Insulin Resistance (HOMA)
Chinese (n= 10)
Caucasian (n= 20) 20
3.0 15

2.5 HOMA
2.0 10
1.0 5
0.0 0
0 15 30 45 60 75 90 105 120 SE Asian Indian Arabic Chinese Caucasian
Time (minutes) n = 10 n = 10 n = 10 n = 10 n = 10

Fig. 5.2 Lean Chinese and Southeast Asians (open studies compared to their Europid counterparts (black
circle) with normal glucose tolerance exhibited higher circle), supporting their biological vulnerability to
glucose excursion during a 75 g oral glucose tolerance test develop diabetes under conditions of metabolic stress [18]
and reduced insulin sensitivity during euglycaemic clamp

Genomic Landscape of Diabetes generations [19] may contribute to this Asian epi-
in the Region demic with increasingly young age of onset. On
the other hand, monogenic diabetes and latent
Most of the genetic variants discovered in the autoimmune diabetes in adults [25] are not uncom-
genome-wide association studies (GWAS) in mon in young Asian populations which further
Caucasians have been replicated in Asian popula- increase the genotypic and phenotypic heterogene-
tions, albeit often with subtle differences in loca- ity of diabetes in Asian populations [26] (Fig. 5.3).
tions and frequency of genetic variants given the In a recent epigenome-wide association study
same locus [19] and larger effect size on beta-cell of South Asians and Caucasians, the methylation
function in Asians for the same variant [20]. levels of several genetic loci associated with lipid
Besides, GWAS performed in Japanese, Chinese metabolism and beta-cell biology were found to be
and Caucasian populations have revealed signifi- associated with increased risk of diabetes in a
cant interethnic differences in genomic architec- dose-dependent manner [27]. These findings are in
ture as well as novel genetic variants in Asian line with the preponderance of noncoding variants
populations which are implicated in protein syn- linked to diabetes in GWAS, suggesting that dys-
thesis and developmental and beta-cell biology regulation of expression may be more important
[19, 21, 22]. Some of these novel variants were than protein changes as a genetic cause for the
first discovered in patients with familial young- common form of diabetes [28]. In this light, mater-
onset diabetes, suggesting that more variants may nal obesity and gestational diabetes may influence
be discovered in these populations [23, 24]. These the in utero environment to cause dysregulation of
findings also support the notion that insufficient expression of interlinking biological networks
beta-cell response to rapid weight gain with through DNA methylation and chromatin modifi-
progressive personal affluence through multiple cation and contribute towards the high prevalence
5 Diabetes in China and the Western Pacific Region 67

Fig. 5.3 A schematic

diagram showing the
phenotypic and
genotypic heterogeneity Non-obese T2D with
of diabetes in Asian genetic variants
populations with and other risk factors
complex interactions Atypical T2D T1D with or
between genetic, Non-diabetic with uncommon without typical
autoimmune and subjects with or variants presentation
external risk factors without common
genetic variants Monogenic

Obese T2D common

genetic variants and other risk

Not to scale
T1D-type 1diabetes
T2D-type 2 diabetes

of childhood metabolic syndrome [29] and young- Amongst these risk factors, low-grade inflam-
onset diabetes in Asian population [30, 31]. In this mation [48] associated with endemic infections
context, the thrifty genotype or phenotype theory such as chronic hepatitis B viral infections
which states that a trait conducive to survival dur- affecting over 10 % of the Asian population [49,
ing energy-scare condition may become maladap- 50] as well as beta-cell dysfunction associated
tive in an energy-abundant environment may be with the use of tobacco, with smoking rates as
particularly relevant to emerging economies like high as 50 % amongst Asian men from China,
China and the WP Region [32]. In support of this Japan, Korea and Indonesia [51], may be partic-
notion, early-life exposure to the 19501960 fam- ularly relevant to the Asian environment. These
ine in China was found to be associated with external factors may interact with cultural fac-
increased risk of diabetes and metabolic syndrome tors such as high consumption of rice with high
[33] especially in subjects with low birth weight glycaemic index [52] to cause metabolic stress
and adult obesity [34, 35]. especially in those with compromised beta-cell
function. Adding to this multicausality and com-
plexity, environmental pollutants such as persis-
Nature Versus Nurture tent organic pollutants (POPs) [53] and bisphenol
A (BPA) [54] are endocrine or mitochondrial
The combination of low BMI and obesity has put disruptors which tend to hit emerging economies
Asians at high risk of diabetes although many of and/or subjects with low socioeconomic status
these risk factors are modifiable [36]. First and hardest. By inducing insulin resistance and/or
foremost, health illiteracy and social disparity beta-cell dysfunction, these chemicals may
[37] remain the most important socioeconomic increase the risk of obesity and hyperglycaemia
determinant for diabetes and obesity [38, 39]. In especially in subjects with genetic predisposi-
these subjects, lack of awareness, poor education, tion [54].
psychosocial stress [40, 41] and long working
hours [42] may amplify the adverse effects of
behavioural risk factors such as poor sleep Morbidity and Mortality
hygiene [43] and the use of tobacco [44], alcohol
[45] and sugar-sweetened beverages [46] which According to the estimates by IDF in 2013, 36 %
can unmask diabetes especially in those with of global deaths in adults are due to diabetes
genetic predisposition [47]. affecting 5.1 million people. The most populous
68 J.C.N. Chan et al.

WP Region had the highest number of diabetes- foot care (11 %) [60] are the main reasons for
related deaths affecting more men (1,080,000) amputation.
than women (789,000). Of note, 44 % of deaths On the other hand, several local, national and
due to diabetes occurred in people under the age regional registries have provided insights regard-
of 60 [1]. On average, diabetes reduces life ing the interethnic differences in diabetic compli-
expectancy by 612 years, especially in people cations and the impact of ageing and care
diagnosed young, and increases the risk of vascu- provision on the secular changes of these clinical
lar, cancer, nonvascular, non-cancer deaths by outcomes. Apart from renal failure, stroke is a
1.33-fold. The nonvascular, non-cancer-related major disease burden in Asian diabetic popula-
deaths are mainly due to renal failure, mental ill- tions [61]. Depending on the demographic pat-
nesses, hepatobiliary disease and sepsis with fast- tern and stage of societal transition, cancer is now
ing plasma glucose having linear relationships emerging as a leading cause of death especially
with all clinical outcomes, even after adjustment in ageing societies and areas with high rate of
for confounders such as age, sex, BMI and life- endemic infections (e.g. hepatitis B for liver can-
style factors [55]. cer) and high survival rates from cardiovascular-
In the WP Region, in contrast to Europids in renal complications [62]. In the Australian
whom cardiovascular disease is a leading cause National Diabetes Registry of nearly one million
of mortality and morbidity, diabetic kidney dis- subjects, researchers have reported 1.3-fold
ease is a major complication in Asia [26], affect- increased standardised incidence risk of all-site
ing 5060 % of type 2 diabetic patients in cancer in both type 1 and type 2 diabetes except
different clinic settings [56]. In prospective stud- for prostate and melanoma [63]. In a national
ies, the annual incidence of diabetic kidney dis- insurance database from Taiwan, diabetes-related
ease in Chinese population was 24 %, driven death rates have declined over time (men, women:
primarily by disease duration, smoking, meta- 3.92 %, 3.29 % in 2000; 3.64 %, 3.11 % in 2005;
bolic syndrome, blood pressure and glycaemic and 3.12 %, 2.71 % in 2009). In 2009, the esti-
control [57]. Other factors such as the use of mated loss of life due to diabetes was 6.1 years in
over-the-counter medications including herbal women and 5.3 years in men amongst people
medicine [58] and chronic hepatitis B infection diagnosed at the age of 40. The four major causes
[50] might also contribute to this high rate of dia- of death were diabetes, malignancies, heart dis-
betic kidney disease in the WP Region. While ease and cerebrovascular disease [64].
many patients in the WP Region die from end- In the Hong Kong Diabetes Registry estab-
stage renal disease due to lack of access to renal lished as a quality improvement programme since
replacement therapy, this devastating condition 1995, researchers were able to track the secular
can be prevented by intensive control of all risk changes in clinical outcomes in Chinese adults
factors using a protocol-driven and collaborative with diabetes, captured by a territory-wide clinical
approach [59]. management system. In the early 1990s when uni-
Countries in the WP Region differ consider- versal health-care coverage was still being devel-
ably in the ecosystem of health-care provision oped, stroke and end-stage renal failure were the
and health literacy. In low-income areas such as leading causes of death. With ageing, societal
some Pacific Islands where health illiteracy is affluence and access to interventions, coronary
common and access to care is poor, leg amputa- heart disease, cancer and heart failure became the
tion continues to be a major cause of morbidity leading causes of death. In 2005 and after a follow-
and mortality. In a recent survey of amputations up period of 5.5 years, amongst the 7534 type 2
involving 85 people with diabetes from diabetic patients enrolled in the registry since
Solomon Islands, Nauru and Vanuatu, delayed 1995, 763 died with the main causes of death being
treatment (42 %), the use of traditional treat- neoplasms (24.5 %) and cardiovascular (23.5 %),
ments (18 %) and insufficient knowledge about respiratory (15.3) and renal disease (15.3 %) [65].
5 Diabetes in China and the Western Pacific Region 69

200 2007 2039 years 2014 2039 years 2007 4059 years
No. with diabetes
150 2014 4059 years 2007 6079 years 2014 6079 years



Africa Europe MENA North SACA Southeast Western World
Amercia Asia Pacific

2039 years 4059 years 6079 years
Percentage change




Africa Europe MENA North SACA Southeast Western World

-40 Amercia Asia Pacific

Fig. 5.4 Rates of increase and number of people with from IDF Atlas 2015 [1]) (MENA Middle East and North
diabetes in different age groups showing the growing epi- Africa, SACA South America and Central America)
demic of young-onset diabetes especially in Asia (Adapted

Emerging Epidemic of Young-Onset implications on the individual, family and society

Diabetes and Its Comorbidities [66]. Between 2007 and 2014, the number of
people with diabetes worldwide has increased by
In the WP Region, 32,500 children under the age 74 % in the 2039 years age group, from 38 mil-
of 15 were estimated to have type 1 diabetes, lion to 67 million. This compares to a 63 %
with the largest number living in the Philippines increase in the 4059 years age group and a 43 %
(7900), followed by China (7700), while increase in the 6079 years age group with an
Australia has the highest estimated incidence rate especially sharp increase in the WP Region and
of 22.3 cases per 100,000 children. In 2013, 5300 Southeast Asia where China and India are located
children were diagnosed to have type 1 diabetes (Fig. 5.4).
in the WP Region [1]. Although type 1 diabetes is In the same registry, the researchers reported
often considered a rare disease in non-Caucasian the rarity of classical type 1 diabetes which
populations, the rising prevalence of young-onset accounted for 3 % in the entire cohort and less
type 2 diabetes is a major health-care challenge. than 10 % amongst those diagnosed before the
Arbitrarily defined as age of diagnosis less than age of 40. Overall, one in five adults with type 2
40 years, these youths and young adults are diabetes had young-onset disease, with 50 %
high-risk populations for premature noncommu- higher risk of cardiovascular-renal disease than
nicable disease (NCD) including but not limited their late-onset counterparts at any given age,
to cardiovascular-renal disease and cancer. This mainly driven by long disease duration [67].
growing epidemic of young-onset diabetes is in Amongst these young-onset type 2 diabetic
part driven by the high prevalence of maternal patients, 40 % were lean and 60 % were obese or
obesity, gestational diabetes and childhood met- overweight. Compared to their counterparts with
abolic syndrome which sets up a vicious cycle type 1 diabetes, these young-onset type 2 dia-
of diabetes begetting diabetes with enormous betic patients had considerably higher rates of
70 J.C.N. Chan et al.

cardiovascular-renal complications, often due to for the prevention and control of type 2 diabetes
poor control of risk factors, frequent default and last updated in 2010 has not yet endorsed the use
treatment non-adherence. By contrast, due to of HbA1c as an alternative to glucose-based tests
their propensity to develop an acute ketotic pre- for diagnosing diabetes [71]. Similarly, the
sentation, type 1 diabetic patients were less likely UNITE FOR Diabetes Philippines in its 2014
to default, and with access to education and insu- practice guideline also recommended against
lin, these patients tended to have low incidence of using A1c. They also proposed the use of OGTT
complications despite long disease duration [68]. for diabetes screening only in individuals with
In a regional database established through the known impaired fasting glucose or metabolic
web-based Joint Asia Diabetes Evaluation syndrome [72]. In the latest report published in
(JADE) programme which enables care providers 2010, the Japan Diabetes Society adopted the use
to establish registries using a common protocol in of A1c to complement glucose-based tests in dia-
a real-world setting, high prevalence of young- betes screening but stipulated that an individual
onset diabetes (~20 %) and treatment gaps was meeting the A1c threshold of 6.5 % must concur-
confirmed in 41,029 adult patients from over ten rently fulfil one other criteria of either fasting
countries in Asia. These real-world data highlight plasma glucose 7.0 mmol/L or 2-h post-OGTT
the heterogeneity in terms of attainment of treat- plasma glucose 11.1 mmol/L to establish the
ment targets often due to differences in health- diagnosis [73]. This differs from international
care coverage and financial support for essential guidelines which assign similar weights to
laboratory investigations and medications as key glucose-based studies and A1c in the diagnostic
components for diabetes care. Yet irrespective of schema and allows the diagnosis to be made on
these diversities, in all countries, there was a con- the basis of A1c measurement alone. The Korean
sistent trend that patients with young-onset dia- Diabetes Association fully applied the ADA cri-
betes were less likely to attain treatment targets teria for diagnosing diabetes since 2011 [74].
for blood glucose and blood lipids and were less Glycated haemoglobin has a number of advan-
likely to be prescribed organ-protective drugs tages over plasma glucose in that it summarises
such as statins and renin-angiotensin system long-term glucose exposure, has less intra-
(RAS) inhibitor despite having clear indications individual variability and does not require testing
such as cardiovascular-renal complications [30]. during a fasted state [75]. The recommendation
to include A1c in diagnostic criteria of diabetes
by the international expert committee was based
Diagnosis of Diabetes on epidemiological studies which demonstrated a
and Prediabetes in Asia tight association between A1c and diabetic reti-
nopathy in mixed populations [7577]. Progress
The criteria for diagnosing diabetes and predia- in analytic procedures and global movement to
betes in Asia for the most part take references certify A1c assay to the National Glycohemoglobin
from international guidelines including those of Standardization Program facilitated its wider use
the World Health Organization (WHO) and as a diagnostic tool as well as in disease monitor-
American Diabetes Association (ADA) [69, 70]. ing [75, 78]. In developing countries such as
The measurement of fasting plasma glucose for China and the Philippines, however, the measure-
diabetes screening is widely accepted and prac- ment of A1c has not yet achieved standardisation
tised. On the other hand, there are differences in across the nation, and this remains the key limita-
regional recommendations for oral glucose tol- tion to its application [72].
erance test (OGTT) and glycated haemoglobin Notwithstanding logistics of costs and instru-
(A1c) in the routine diagnostic process. mentation, inclusion of A1c in diagnostic crite-
Approaches to screening are also closely related ria will uncover more patients with diabetes than
to health-care access and resources available. For using glucose-based criteria alone. In large pop-
instance, the Chinese Diabetes Society guideline ulation surveys in China and Japan, over half of
5 Diabetes in China and the Western Pacific Region 71

the participants with newly diagnosed diabetes ducted in Asia which showed that fasting plasma
and prediabetes were detected with A1c. This glucose alone could only identify half of the peo-
may be due to the high prevalence of high post- ple with diabetes and that individuals with pre-
glucose challenge blood glucose levels which dominantly post-load hyperglycaemia are
may be detected by an integrated value like A1c phenotypically different from those with fasting
but missed by using fasting plasma glucose only hyperglycaemia [86]. As discussed, Asian dia-
[4, 79]. betic patients have lower BMI and less beta-cell
Increasingly, it is recognised that ethnicity reserve than their Caucasian counterparts and
may modify the relationship between measured thus are more likely to fail glucose stress test. On
A1c and blood glucose levels. Several studies this premise, a lower threshold to perform OGTT
have shown that for similar distribution of blood in Asians should be considered. In Chinese, a
glucose, A1c levels were consistently higher in paired value of A1c 6.1 % and fasting plasma
Asians compared to Caucasians and that amongst glucose 6.1 mmol/L had 1317 times increased
Asians, higher in Indians and Malays relative to likelihood to occur in diabetic subjects compared
Chinese [80, 81]. A few factors may contribute to to nondiabetic subjects [87, 88]. In resource-
such racial disparities such as differences in hae- limited setting, OGTT will have the greatest yield
moglobin glycation rate and in population preva- in detecting diabetes when paired A1c and fasting
lence of haemoglobinopathies which interfere plasma glucose are above these cut-off values.
with A1c measurement.
Against this background, concerns are raised
regarding the optimal A1c threshold for detecting Translating Primary Prevention
diabetes in Asian groups and whether ethnic- of Type 2 Diabetes
specific standards are required. Studies carried
out in Asians have suggested that lower cut-offs The chronic, silent and nonurgent nature of dia-
of A1c correlated more closely with diabetes as betes often leads to late diagnosis and delayed
defined using the gold standard of OGTT [82 treatment. Once diagnosed, people with diabetes
84]. In a recent population-based study compris- have pluralistic needs including information,
ing three ethnic groups of Malay, Chinese and social and psychological support, education,
Indian which examined the cross-sectional rela- medical and surgical treatment, to name but a
tionship between A1c and moderate retinopathy, few. Given these multiple demands, large popula-
it was found that although A1c differed slightly tion size, lack of capacity and resource restraints,
in its sensitivity and specificity in predicting reti- prevention and control of diabetes is often chal-
nopathy amongst different ethnic groups, eye lenged with care fragmentation, clinical inertia
complication was generally rare when A1c fell and treatment non-adherence with insufficient
below 6.5 % [85]. Results from this Asian study community support [3, 89, 90]. In both ran-
concurred with those of a large multinational domised disease management programmes and
database which confirmed that an A1c threshold observational cohorts, attaining multiple treat-
of 6.5 % has similar discriminatory power for the ment targets and the use of organ-protective
presence or absence of diabetic retinopathy in drugs such as statins and RAS inhibitors have
Caucasians and Asians [77, 85]. been shown to reduce cardiovascular-renal com-
Oral glucose tolerance test is the gold standard plications and related deaths in the WP Region
test for the diagnosis of diabetes and is necessary [91, 92]. Yet large-scale surveys have demon-
in making a diagnosis of impaired glucose toler- strated only 30 % of patients with diabetes ever
ance (IGT). The requirement of a second blood had assessments for risk factors and complica-
sample and being more time consuming has tions. Amongst those who had assessments,
greatly limited its practice in many Asian coun- only 3040 % attained one of the three ABC
tries. However, the importance of the 2-h plasma targets (A1c <7 %, BP <130/80 mmHg, LDL-C
glucose has been highlighted in studies con- <2.6 mmol/L) and 510 % attained all three
72 J.C.N. Chan et al.

targets [93] with particularly low rates amongst strategic alliance between the WHO Regional
people with young-onset diabetes [30, 94]. Office for the WP, the Secretariat of the Pacific
Diabetes is a lifelong disease which requires Community and the IDF-Western Pacific Region
acquisition of knowledge, change of attitudes and as a regional voice to advocate for prevention and
formation of new habits in order to reduce dis- control of diabetes. In the WPDD Plan of Action
ease burden. It is also a biological problem that (20052010), experts from the region sum-
requires medications, technologies and monitor- marised the multidimensional nature of diabetes
ing which have resource implications. In a survey (Fig. 5.5) and the need to use a multipronged
from Guam involving 125 patients, 40 % were strategy including policies and legislations
not aware of the type of diabetes they had, 20 % (Fig. 5.6) to create a health-enabling environment
had not received education on self-management and reform the health-care system to make pre-
and 3060 % had not received advice on tobacco vention and control of diabetes accessible, afford-
cessation, immunisation services, nutritional able and sustainable [101].
counselling and/or regular eye and foot examina-
tions. Over 50 % of patients expressed their desire
to have preventive and education services with Some Examples of Diabetes
enhanced access to specialists and specialised Prevention and Control Strategies
care as well as financial support to cover the costs from China and WP
of chronic care and medications [95].
There is now consensus that a chronic care National Plan for Diabetes and NCD
model involving community mobilisation and Prevention
engagement supported by an integrated health- In the China National Plan for NCD Prevention
care system with country/area appropriate financ- and Treatment (20122015), the government pro-
ing and health-care organisation is needed to posed to use public measures including environ-
promote interactions between an informed per- mental protection, food safety, clean drinking
son with diabetes and a proactive health-care water, tobacco control and occupational safety to
team to achieve positive clinical outcomes [96]. create a health-enabling environment through
In this information and technology era, there is a broad social participation and multisector collab-
growing advocacy to use improvement science, orations. Other proposed practices and policies
often entailing the use of real-world data, to included universal vaccination, maternal and
inform and change clinical practice [97]. In the child health programmes and primary care. Since
case of diabetes, changing the workflow and then, the Chinese government has introduced dif-
using protocols to collect data systematically for ferent health insurance packages at the rural and
establishing registry can generate personalised urban areas and built more than 30,000
report to promote shared decision-making. On a community-based clinics to provide basic preven-
broader perspective, data transparency and com- tive care programme including regular risk factor
parison of performance indexes may also moti- surveillance and provision of essential medica-
vate practice and policy changes through ongoing tions for control of blood pressure and blood glu-
evaluation and sharing of best care models cose [102]. Despite these ongoing efforts, given
[98100]. the vastness of the country and size of the popula-
tion, there remain challenges in communications
amongst relevant stakeholders, patient flow
Western Pacic Diabetes Declaration between hospitals and primary care clinics, capac-
(WPDD) Plan for Action (20062010) ity building and public education [103].

In 2000, at the 51st session of the WHO Regional Tobacco Control

Committee for the WP, the Committee endorsed Several meta-analyses have indicated a dose-
the establishment of the WPDD as a regional dependent risk association of smoking with dia-
5 Diabetes in China and the Western Pacific Region 73

Diabetes and cardio-

renal complications

Chronic disease,
Obesity and morbidity and
metabolic syndrome mortality
Personal lifestyle

Physical activity e.g. Unhealthy diet e.g. Smoking,alcohol

TV watching, video soft drink junk food, binge drinking
game playing, low fruit/vegetable
computer use and high meat
Environment and consumption

Socio-economical Employment, income, Demographics and Political environment Other factors e.g. low
status,education, access, to care and culture e.g. aging and policy and grade infections,
psychosocial stress medical insurance migrant populations legislations maternal and
perinatal health, low
birth weight


Fig. 5.5 A schematic diagram summarising the interac- in the Western Pacific Diabetes Declaration Plan for
tions amongst predisposing, precipitating and perpetuat- Action (20052010) [154]
ing factors in the epidemic of diabetes and its comorbidities

Policy makers Health care team

Collection of information (STEPS) at Development of clinical guidelines

periodic intervals

Information dissemination Patient education and self-care

Health awareness Periodic assessments and default
Mass media Treatment to targets
Personnel training
Resource redistribution
New funding model
Nongovernmental organizations
Medical insurance
Other sectors

Health Promotion Programmes Disease management

(diet, exercise, tobacco and alohol) Integrated and continual care

School Screening for Community Hospital

Policy diabetes in high care care
risk subjects

Ongoing data Ongoing data

collection collection

Reduce obesity and diabetes Reduce diabetes complications

Fig. 5.6 A multipronged strategy using policies, commu- Western Pacific Diabetes Declaration Plan for Action
nity mobilisation, intersectoral partnerships and clinical (20052010) [154]
leaderships to detect, prevent and control diabetes in the
74 J.C.N. Chan et al.

betes [104] as well as that with cardiovascular WHO has now called for early prevention of dia-
disease and total mortality in people with diabetes betes using a life-course strategy that starts with
[105]. Although it remains to be confirmed maternal health [116]. However, screening for
whether smoking cessation will reduce the risk of gestational diabetes is only practised in a few
diabetes, it is likely that tobacco control would countries in Asia [117]. Several countries such as
reduce the amplifying effects of tobacco on mor- Japan [118] and Taiwan [119] have embarked
bidities associated with diabetes, notably, cancer upon national urine glucose screening pro-
and cardiovascular-renal disease. To this end, gramme, while a few countries such as Singapore
there have been successful stories in tobacco con- have introduced national school programme to
trol through strong political will and tough mea- control childhood obesity [120].
sures in the WP Region, such as New Zealand and
Hong Kong [106].
Rational Selection of Antidiabetes
Early Detection and Prevention Medications
There is high-level evidence confirming that many
cases of diabetes can be prevented or delayed by In previous sections, we have highlighted differ-
structured lifestyle modification [107]. That said, ences in pathophysiology in Asian populations.
most experts and professional bodies proposed tar- Beta-cell dysfunction may be the primary defect
geted or opportunistic screening using known risk in lean individuals, while Asians exhibit greater
factors (e.g. age, sex, family history, obesity (BMI insulin resistance and visceral adiposity at a lower
and waist circumference), blood pressure and ges- BMI compared with Caucasian counterparts.
tational diabetes) and/or simple risk scores [108] Therefore, treatment may be tailored to reflect
rather than population screening [109]. Given the these ethnic differences. Metformin is widely
high prevalence of young-onset diabetes in the WP used as a first-line oral antidiabetic agent. Despite
Region and the importance of disease duration as a a lower BMI, metformin appears efficacious in
determinant for all diabetes-related complications Asian populations [121]. Asian diabetic subjects
[110], targeted screening in young subjects with treated with metformin show a similar reduction
family history and/or obesity is likely to be cost- in insulin resistance and improved glycaemic con-
effective due to the high prevalence of diabetes trol as compared with non-Asians [122].
amongst the relatives of affected subjects [111] Therefore, metformin, unless contraindicated, is
and the high lifetime risk for complications, once recommended as the first-line therapy according
affected [109]. to global guidelines by the IDF [123].
In the early 1990s, China conducted the first Alpha glucosidase inhibitors (AGI) may have a
randomised 6-year lifestyle modification pro- particular role in early therapy amongst Asians
gramme in a workforce who had IGT. These sub- due to high carbohydrate content of Asian diets.
jects continued to be observed for more than The typical Chinese diet has a carbohydrate con-
20 years after completion of the trial which con- tent of up to 67 % as compared with 40 % amongst
firmed that early intervention reduced the inci- Europeans [124]. In one study, treatment with
dence of diabetes, retinopathy and cardiovascular acarbose 50100 mg three times a day effectively
disease [112, 113]. Similar prevention pro- reduced A1c by 0.7 % and improved both fasting
grammes have also been successfully translated and 1 h plasma glucose as compared with placebo
and adapted to meet the special needs in primary in Asians with type 2 diabetes [125]. In the met-
care settings in Australia [114] and indigenous formin and acarbose in Chinese as the initial
populations in Pacific Islands [115]. hypoglycaemic treatment (MARCH) study,
The negative impact of gestational diabetes on nearly 800 newly diagnosed Chinese patients with
childhood obesity may underlie the growing epi- type 2 diabetes were randomised to 24 weeks of
demic of young-onset diabetes and NCD. The either acarbose or metformin monotherapy [126].
5 Diabetes in China and the Western Pacific Region 75

The reduction in A1c was comparable in the acar- to-head trials comparing glinides versus other
bose- and metformin-treated groups (1.17 % first-line oral antidiabetic drugs; therefore, their
versus 1.19 %), demonstrating noninferiority in role in early therapy remains unclear.
overall glucose lowering. In contrast to Caucasians, the incretin response
Further, there is a strong evidence that AGI is preserved in Asians with type 2 diabetes [131].
may prevent progression from IGT to type 2 dia- DPP4 inhibitors may have a larger effect in glucose
betes. In the STOP-NIDDM trial that was con- reduction in Asians as compared with non-Asians.
ducted in Canada and Europe, the use of acarbose In a systematic review of randomised controlled
reduced progression to type 2 diabetes by 25 % trials of DPP4 inhibitors, those involving higher
[127]. These findings have been replicated in the proportion of Asian subjects demonstrated greater
VICTORY study which compared voglibose A1c lowering [132]. Similarly, GLP-1 analogues
versus placebo amongst 1780 Japanese individu- lowered A1c to a greater extent in Asian-dominant
als with IGT. The use of voglibose was associ- as compared with non-Asian-dominant studies
ated with an impressive 40 % lower risk of [133]. Although there are no ethnic-specific data of
progression to type 2 diabetes [128]. The control long-term safety of incretin-based therapies, sev-
of postprandial glucose excursions may theoreti- eral large-scale cardiovascular end point studies of
cally confer cardioprotection via reduction of saxagliptin and sitagliptin have enrolled over 10 %
oxidative stress. However, the jury is still out as of Asians and demonstrated neutral effects on car-
to whether AGI confers any cardiovascular ben- diovascular outcomes [134, 135].
efit. In an analysis of a large Taiwanese cohort Severe beta-cell dysfunction is common at
comparing type 2 diabetes patients treated with diagnosis, and there is evidence that early inten-
AGI versus metformin monotherapy, the risk of sive insulin therapy may preserve beta-cell func-
cardiovascular events was significantly higher tion or even induce remission in Asians. An initial
amongst those on AGI, adjusted for baseline dif- study showed that continuous subcutaneous insu-
ferences by propensity score [129]. Ongoing tri- lin infusion for 2 weeks in newly diagnosed type
als, such as the Acarbose Cardiovascular 2 diabetes Chinese patients led to restoration of
Evaluation (ACE) study which evaluates the early-phase insulin secretion and beta-cell func-
effect of acarbose on recurrent cardiovascular tion. As such, 42 % of patients maintained optimal
events and incident diabetes amongst those with glycaemic control after 2 years on diet alone
history of cardiovascular disease and IGT, may [136]. A subsequent study led by Weng and col-
provide much needed answers [130]. leagues randomised 410 newly diagnosed type 2
Insulin secretagogues including sulfonylureas diabetes patients to multiple daily injections
and glinides are widely used and are generally (MDI), continuous subcutaneous insulin infusion
efficacious in glucose lowering amongst Asians. (CSII) or oral agents, and treatment was stopped
The glinides target early-phase insulin release, as after normoglycaemia was reached for 2 weeks
compared with sulfonylureas which tend to have a [137]. The study demonstrated superior remission
longer half-life. Thus, glinides may be particu- rates at 1 year in insulin-treated patients regard-
larly effective in Asian Chinese where a defect in less of route (CSII 51 %, MDI 45 %) as compared
early-phase insulin is common [124]. In a trial in with those treated with oral agents (27 %). Similar
Chinese patients which compared repaglinide levels of A1c reduction were achieved in all three
plus metformin versus repaglinide alone, both groups, suggesting the restoration of beta-cell
arms achieved a similar reduction in A1c (4.45 % function is not merely linked to amelioration of
with repaglinide plus metformin versus 4.15 % hyperglycaemia. These studies argue for a possi-
with repaglinide only) from baseline A1c of 10 %. ble use of intensive insulin therapy to reverse
However, combination treatment with metformin acute glucotoxicity for preserving beta-cell func-
achieved lower fasting plasma glucose as com- tion in Asian patients especially in those with
pared with repaglinide alone. There are no head- early diabetes and without complications.
76 J.C.N. Chan et al.

Given the phenotypic heterogeneity, subtle dif- patients who need additional informational, medi-
ferences in disease mechanisms and lifestyle fac- cal and social support (Fig. 5.7) [140].
tors, the goal of a personalised regimen is to By combining service provision and epide-
match the right drug to the right patient at the miological research, these cohorts, databases
right time for the right outcome. In particular the and biobanks have provided clinicians unique
risk-benefit of each therapeutic agent should be opportunities to track clinical progress, develop
considered for in the individual in order to achieve risk equations and identify treatment gaps such
the best outcomes [138]. as the associations between hypoglycaemia, can-
cer and diabetic kidney disease [141] as well as
interactions amongst genotypes, phenotypes,
Coordination and Delivery treatment and clinical outcomes [142, 143]. In
of Diabetes Care Services 2009, the Hong Kong government fully adopted
this integrated care model and introduced the
In a meta-analysis, quality improvement initia- risk assessment management programme and
tives targeting at patients, systems and care pro- patient empowerment programme to the primary
viders have been shown to reduce blood pressure, care clinics which were shown to improve con-
blood glucose and blood lipids although amongst trol of all risk factors and reduce cardiovascular
these strategies, patient education, task delega- disease and all-cause death [144146].
tion, self-care and case management were found
to have the largest effect sizes [99]. Hong Kong is
one of the most cosmopolitan cities in China with Using the JADE Registry to Promote
a population of seven million. It has a public Quality Assurance and Cooperative
health-care system delivered through the Hospital Learning
Authority which funds and governs all public hos-
pitals and clinics, modelled after the UK National In 2007, this concept of using structured processes
Health Scheme. to personalise care [147], augmented by informa-
In response to the growing epidemic of diabetes tion technology and collaborative care [148], was
since the early 1990s and given limited resources, digitalised through the establishment of the web-
clinical researchers and care providers used knowl- based Joint Asia Diabetes Evaluation (JADE) pro-
edge transfer and changed care settings to cope gramme using private-public partnerships [139].
with the patient volume. By setting up a hospital- By joining the JADE programme, practitioners can
based diabetes centre, located away from the busy establish their own clinic registry using built-in
clinics, coordinated by specialised nurses and templates and protocols for quality assurance while
health-care assistants and supported by endocri- contributing anonymous data to form the JADE
nologists, the diabetes team offers regular risk Registry to identify treatment gaps and share best
assessment and education services to diabetic practices [139]. Using this regional registry, first of
patients referred by all specialties. A key feature of its kind, the researchers were able to confirm the
this model is the use of protocols, task delegation high prevalence of young-onset diabetes (20 %) in
and data management to establish the Hong Kong Asia and the treatment gaps [30] as well as the ben-
Diabetes Registry for risk stratification and quality efits of providing informational and social support
assurance [139]. These data were used to generate on reducing clinical inertia, treatment non-adher-
personalised report for improving communica- ence and hospitalisations [140, 149].
tions amongst specialists, doctors and patients as
well as triaging patients into primary or hospital-
based care. The Diabetes Center also serves as a Future Directions
central point for coordinating various activities
including peer support, group medical follow-up In 2009, the United Nations General Assembly
and behavioural programmes for difficult-to-treat identified diabetes, cardiovascular disease, cancer
5 Diabetes in China and the Western Pacific Region 77

Comminity Diabetes Public health

Empowerment assessment Measures & policy
& education


Treat early,
Case safely and to Information
finding multiple technology


Resourcing Health literacy

Fig. 5.7 A conceptual framework of using diabetes cen- linked by information technology and protocols to estab-
tres run by paramedical staff, supported by internists, to lish registry for quality assurance. Through community
provide assessment, education, outreach and peer support empowerment, capacity building and health-care policies,
programmes to complement medical care provided by these centres can contribute towards early detection, pre-
specialists and family doctors. All parties concerned are vention and control of diabetes and its comorbidities

and respiratory disease as the four NCDs as top pri- centres for education to develop, produce and dis-
orities for prevention and control [150]. Recently, seminate diabetes education in order to enhance
universal health coverage is ranked a top WHO professional education for which two of the nine
agenda for ensuring people with NCD to have centres were located in the WP Region (Australia
access to medications, technologies and basic and Hong Kong) [152].
medical care [151]. Given the diversity of the WP
Region, different countries will have their unique Conclusion
transition patterns in terms of political, socio-eco- Although the definition of health by the
nomical, technological, cultural and health-care WHO as a state of complete physical, mental
development which will interact to influence the and social well-being and not merely the
diabetes landscapes in these countries. As such, absence of disease or infirmity [153] may not
each country/area will have to develop its own be easily attained, collective efforts through
prevention and control strategies tailored to its government mandates, clinical leaderships,
special needs and circumstances to meet the man- intersectoral partnerships, community
dates set by the WHO and UN Assembly. empowerment, ongoing research and advo-
Policies aside, from a more operational per- cacy and prevention and control of diabetes
spective, health-care providers who are equipped are aspiration that is feasible, achievable and
with knowledge about the nature of diabetes and indeed essential. To this end, with increasing
stand between evidence and patients in need of globalisation, more countries/areas will
evidence-based care, knowledge transfer and undergo urbanisation and as such, the expo-
application become their armamentarium in the sures to, and experiences gained with diabetes
pursuit of health protection and disease preven- in the WP Region, will provide invaluable les-
tion. In this regard, the IDF, as a global voice for sons in our common pursuit of prevention and
people with diabetes, has established reference control of diabetes and its comorbidities.
78 J.C.N. Chan et al.

References 17. Yoon KH, Ko SH, Cho JH, et al. Selective beta-cell
loss and alpha-cell expansion in patients with type 2
diabetes mellitus in Korea. J Clin Endocrinol Metab.
1. IDF Atlas 2015: International Diabetes Federation
18. Dickinson S, Colagiuri S, Faramus E, Petocz P,
2. Chan JC, Cho NH, Tajima N, Shaw J. Diabetes in the
Brand-Miller JC. Postprandial hyperglycemia and
Western Pacific Region-past, present and future.
insulin sensitivity differ among lean young adults of
Diabetes Res Clin Pract. 2014;103:24455.
different ethnicities. J Nutr. 2002;132:25749.
3. Chan JC, Zhang Y, Ning G. Diabetes in China: a
19. Ma RC, Chan JCN. Type 2 diabetes in East Asians:
societal solution for a personal challenge. Lancet
similarities and differences with populations in
Diabetes Endocrinol. 2014;2:96979.
Europe and the United States. Ann N Y Acad Sci.
4. Xu Y, Wang L, He J, et al. Prevalence and control of
diabetes in Chinese adults. JAMA J Am Med Assoc.
20. Hong KW, Chung M, Cho SB. Meta-analysis of
genome-wide association study of homeostasis model
5. DeCODA Study Group. Prevalence of the metabolic
assessment beta cell function and insulin resistance in
syndrome in populations of Asian origin.
an East Asian population and the European results.
Comparison of the IDF definition with the NCEP
Mol Genet Genom MGG. 2014;289:124755.
definition. Diabetes Res Clin Pract. 2007;76:5767.
21. Cho YS, Chen CH, Hu C, et al. Meta-analysis of
6. Stevens GA, Singh GM, Lu Y, et al. National,
genome-wide association studies identifies eight
regional, and global trends in adult overweight and
new loci for type 2 diabetes in East Asians. Nat
obesity prevalences. Popul Health Metrics. 2012;
Genet. 2011;44:6772.
22. Yamauchi T, Hara K, Maeda S, et al. A genome-
7. Khambalia A, Phongsavan P, Smith BJ, et al. wide association study in the Japanese population
Prevalence and risk factors of diabetes and impaired identifies susceptibility loci for type 2 diabetes at
fasting glucose in Nauru. BMC Public Health. 2011; UBE2E2 and C2CD4A-C2CD4B. Nat Genet. 2010;
11:719. 42:8648.
8. International Diabetes Federation, ed. International 23. Ma RC, Lee HM, Lam VK, et al. Familial young-
Diabetes Atlas 2013. onset diabetes, pre-diabetes and cardiovascular dis-
9. Hu FB. Globalization of diabetes: the role of diet, ease are associated with genetic variants of
lifestyle, and genes. Diabetes Care. 2011;34: DACH1 in Chinese. PLoS ONE. 2014;9, e84770.
124957. 24. Ma RC, Hu C, Tam CH, et al. Genome-wide associa-
10. Dancause KN, Dehuff C, Soloway LE, et al. tion study in a Chinese population identifies a sus-
Behavioral changes associated with economic devel- ceptibility locus for type 2 diabetes at 7q32 near
opment in the South Pacific: health transition in PAX4. Diabetologia. 2013;56:1291305.
Vanuatu. Am J Hum Biol Off J Hum Biol Counc. 25. Kong APS, Chan JCN. Other disorders with type 1
2011;23:36676. phenotype. In: Holt R, Goldstein B, Flyvbjerg A,
11. Ma RC, Lin X, Jia W. Causes of type 2 diabetes in Cockram CS, editors. Textbook of diabetes. 4th ed.
China. Lancet Diabetes Endocrinol. 2014;2:98091. Chichester: Wiley-Blackwell, Wiley; 2010. p. 1529.
12. Kahn SE. The relative contributions of insulin resis- 26. Kong AP, Xu G, Brown N, So WY, Ma RC, Chan
tance and beta-cell dysfunction to the pathophysiol- JC. Diabetes and its comorbidities-where East meets
ogy of Type 2 diabetes. Diabetologia. 2004;46 West. Nat Rev Endocrinol. 2013;9:53747.
:319. 27. Chambers JC, Loh M, Lehne B, et al. Epigenome-
13. Kodama K, Tojjar D, Yamada S, Toda K, Patel CJ, wide association of DNA methylation markers in
Butte AJ. Ethnic differences in the relationship peripheral blood from Indian Asians and Europeans
between insulin sensitivity and insulin response: a with incident type 2 diabetes: a nested case-control
systematic review and meta-analysis. Diabetes Care. study. Lancet Diabetes Endocrinol. 2015;3:52634.
2013;36:178996. 28. Lou S, Lee HM, Qin H, et al. Whole-genome bisul-
14. Deurenberg-Yap M, Schmidt G, van Staveren WA, fite sequencing of multiple individuals reveals com-
Deurenberg P. The paradox of low body mass index plementary roles of promoter and gene body
and high body fat percentage among Chinese, Malays methylation in transcriptional regulation. Genome
and Indians in Singapore. Int J Obes. 2000;24:10117. Biol. 2014;15:408.
15. King GL, McNeely MJ, Thorpe LE, et al. 29. Ma RC, Tutino GE, Lillycrop KA, Hanson MA, Tam
Understanding and addressing unique needs of dia- WH. Maternal diabetes, gestational diabetes and the
betes in Asian Americans, native Hawaiians, and role of epigenetics in their long term effects on off-
Pacific Islanders. Diabetes Care. 2012;35:11818. spring. Prog Biophys Mol Biol. 2015;118:5568.
16. Suraamornkul S, Kwancharoen R, Ovartlarnporn M, 30. Yeung RO, Zhang Y, Luk A, et al. Metabolic profiles
Rawdaree P, Bajaj M. Insulin clamp-derived mea- and treatment gaps in young-onset type 2 diabetes in
surements of insulin sensitivity and insulin secretion Asia (the JADE programme): a cross-sectional study
in lean and obese Asian type 2 diabetic patients. of a prospective cohort. Lancet Diabetes Endocrinol.
Metab Syndr Relat Disord. 2010;8:1138. 2014;2:93543.
5 Diabetes in China and the Western Pacific Region 79

31. Chan JC, Yeung R, Luk A. The Asian diabetes phe- style factors in Chinese: the Better Health for Better
notypes: challenges and opportunities. Diabetes Res Hong Kong health promotion campaign. Eur J Clin
Clin Pract. 2014;105:1359. Nutr. 2010;64:138692.
32. Diamond JM. Diabetes running wild. Nature. 47. Qi Q, Chu AY, Kang JH, et al. Sugar-sweetened bev-
1992;357:3623. erages and genetic risk of obesity. N Engl J Med.
33. Li Y, Jaddoe VW, Qi L, et al. Exposure to the Chinese 2012;367:138796.
famine in early life and the risk of metabolic syn- 48. Pickup JC. Inflammation and activated innate immu-
drome in adulthood. Diabetes Care. 2011;34:10148. nity in the pathogenesis of type 2 diabetes. Diabetes
34. McCance DR, Pettitt DJ, Hanson RL, Jacobsson Care. 2004;27:81323.
LTH, Knowler WC, Bennett PH. Birth weight and 49. Lao TT, Chan BC, Leung WC, Ho LF, Tse
non-insulin dependent diabetes: thrifty genotype, KY. Maternal hepatitis B infection and gestational
thrifty phenotype, or surviving small baby geno- diabetes mellitus. J Hepatol. 2007;47:4650.
type? Br Med J. 1994;308:9425. 50. Cheng AY, Kong AP, Wong VW, et al. Chronic hepa-
35. Neel JV. Diabetes mellitus: a thrifty genotype ren- titis B viral infection independently predicts renal
dered detrimental by progress? Am J Hum Genet. outcome in type 2 diabetic patients. Diabetologia.
1962;14:35362. 2006;49:177784.
36. Boyko EJ, Fujimoto WY, Leonetti DL, Newell- 51. Ueshima H, Sekikawa A, Miura K, et al.
Morris L. Visceral adiposity and risk of type 2 diabe- Cardiovascular disease and risk factors in Asia: a
tes: a prospective study among Japanese Americans. selected review. Circulation. 2008;118:27029.
Diabetes Care. 2000;23:46571. 52. Yu R, Woo J, Chan R, et al. Relationship between
37. Wong MC, Leung MC, Tsang CS, Lo SV, Griffiths dietary intake and the development of type 2 diabe-
SM. The rising tide of diabetes mellitus in a Chinese tes in a Chinese population: the Hong Kong Dietary
population: a population-based household survey on Survey. Public Health Nutr. 2011;14:113341.
121,895 persons. Int J Public Health. 2013;58:26976. 53. Lim S, Cho YM, Park KS, Lee HK. Persistent
38. Demakakos P, Nazroo J, Breeze E, Marmot organic pollutants, mitochondrial dysfunction, and
M. Socioeconomic status and health: the role of sub- metabolic syndrome. Ann N Y Acad Sci. 2010;1201:
jective social status. Soc Sci Med. 2008;67:33040. 16676.
39. Ko GTC, Chan JCN, Cockram CS. A low socioeco- 54. Bi Y, Wang W, Xu M, et al. Diabetes genetic risk
nomic class is associated with diabetes and meta- score modifies effect of bisphenol A exposure on
bolic syndrome in Hong Kong Chinese. Eur deterioration in glucose metabolism. J Clin
J Epidemiol. 2001;17:28995. Endocrinol Metab. 2016;101:14350.
40. Williams ED, Magliano DJ, Tapp RJ, Oldenburg BF, 55. Seshasai SR, Kaptoge S, Thompson A, et al.
Shaw JE. Psychosocial stress predicts abnormal glu- Diabetes mellitus, fasting glucose, and risk of cause-
cose metabolism: the Australian diabetes, obesity specific death. N Engl J Med. 2011;364:82941.
and lifestyle (AusDiab) study. Ann Behav Med Publ 56. Wu AY, Kong NC, de Leon FA, et al. An alarmingly
Soc Behav Med. 2013;46:6272. high prevalence of diabetic nephropathy in Asian type
41. Harding JL, Backholer K, Williams ED, et al. 2 diabetic patients: the MicroAlbuminuria Prevalence
Psychosocial stress is positively associated with (MAP) Study. Diabetologia. 2005;48:1726.
body mass index gain over 5 years: evidence from 57. Luk AO, So WY, Ma RC, et al. Metabolic syndrome
the longitudinal AusDiab study. Obesity. 2014;22(1): predicts new onset of chronic kidney disease in
27786. 5,829 patients with Type 2 diabetes: a 5-year pro-
42. Ko GT, Chan JC, Chan AW, et al. Association spective analysis of the Hong Kong Diabetes
between sleeping hours, working hours and obesity Registry. Diabetes Care. 2008;31:235761.
in Hong Kong Chinese: the better health for better 58. Wen CP, Cheng TY, Tsai MK, et al. All-cause mor-
Hong Kong health promotion campaign. Int J Obes. tality attributable to chronic kidney disease: a pro-
2007;31:25460. spective cohort study based on 462 293 adults in
43. Cappuccio FP, DElia L, Strazzullo P, Miller Taiwan. Lancet. 2008;371:217382.
MA. Quantity and quality of sleep and incidence of 59. Chan JC, So WY, Yeung CY, et al. Effects of struc-
type 2 diabetes: a systematic review and meta- tured versus usual care on renal endpoint in type 2
analysis. Diabetes Care. 2010;33:41420. diabetes: the SURE study: a randomized multi-
44. Cho NH, Chan JC, Jang HC, Lim S, Kim HL, Choi center translational study. Diabetes Care. 2009;32:
SH. Cigarette smoking is an independent risk factor 97782.
for type 2 diabetes: a four-year community-based pro- 60. Win Tin ST, Gadabu E, Iro G, Tasserei J, Colagiuri
spective study. Clin Endocrinol. 2009;71:67985. R. Diabetes related amputations in Pacific Islands
45. Xu M, Zhou Y, Xu B, et al. Associations of smoking countries: a root cause analysis of precipitating
and alcohol consumption with impaired beta-cell events. Diabetes Res Clin Pract. 2013;100:2304.
function in Chinese men. J Diabetes. 2016;8(3): 61. Morrish NJ, Wang S, Stevens LK, Fuller JH, Keen
43441. H. Mortality and causes of death in the WHO
46. Ko GT, So WY, Chow CC, et al. Risk associations of Multinational Survey of Vascular Diseases in Diabetes.
obesity with sugar-sweetened beverages and life- Diabetologia. 2001;44:S1421.
80 J.C.N. Chan et al.

62. Yang X, Lee HM, Chan JC. Drug-subphenotype globin standardization program: a five-year progress
interactions for cancer in type 2 diabetes mellitus. report. Clin Chem. 2001;47:198592.
Nat Rev Endocrinol. 2015;11:3729. 79. Heianza Y, Hara S, Arase Y, et al. Impact of introduc-
63. Harding JL, Shaw JE, Peeters A, Cartensen B, ing HbA1c into the diagnostic criteria on prevalence
Magliano DJ. Cancer risk among people with type 1 and cardiovascular risk profiles of individuals with
and type 2 diabetes: disentangling true associations, newly diagnosed diabetes in Japan: the Toranomon
detection bias, and reverse causation. Diabetes Care. Hospital Health Management Center Study 2
2015;38:26470. Diabetes care 2015;38:7345. (TOPICS 2). Diabetes Res Clin Pract. 2012;95:
64. Li HY, Jiang YD, Chang CH, Chung CH, Lin BJ, 28390.
Chuang LM. Mortality trends in patients with dia- 80. Venkataraman K, Kao SL, Thai AC, et al. Ethnicity
betes in Taiwan: a nationwide survey in 20002009. modifies the relation between fasting plasma glucose
J Formos Med Assoc Taiwan yi zhi. 2012;111: and HbA1c in Indians, Malays and Chinese. Diabet
64550. Med. 2012;29:9117.
65. So WY, Yang X, Ma RC, et al. Risk factors in 81. Herman WH, Cohen RM. Racial and ethnic differ-
V-shaped risk associations with all-cause mortal- ences in the relationship between HbA1c and blood
ity in type 2 diabetes-The Hong Kong Diabetes glucose: implications for the diagnosis of diabetes.
Registry. Diabetes Metab Res Rev. 2008;24:23846. J Clin Endocrinol Metab. 2012;97:106772.
66. Chan JC, Malik V, Jia W, et al. Diabetes in Asia: epi- 82. Lee H, Oh JY, Sung YA, et al. Optimal hemoglo-
demiology, risk factors, and pathophysiology. JAMA bin A1C cutoff value for diagnosing type 2 diabetes
J Am Med Assoc. 2009;301:212940. mellitus in Korean adults. Diabetes Res Clin Pract.
67. Chan JC, Lau ES, Luk AO, et al. Premature mortal- 2013;99:2316.
ity and co-morbidities in young-onset diabetes a 83. Ko GT, Chan JC, Tsang LW, Cockram CS. Combined
7 year prospective analysis. Am J Med. 2014;127: use of fasting plasma glucose and HbA1c predicts
61624. the progression to diabetes in Chinese subjects.
68. Luk AO, Lau ES, So WY, et al. Prospective study on Diabetes Care. 2000;23:17703.
the incidences of cardiovascular-renal complications 84. Wu S, Yi F, Zhou C, et al. HbA1c and the diagno-
in Chinese patients with young-onset type 1 and type sis of diabetes and prediabetes in a middle-aged
2 diabetes. Diabetes Care. 2014;37:14957. and elderly Han population from northwest China
69. Diagnosis and classification of diabetes mellitus. (HbA1c). J Diabetes. 2013;5:28290.
Diabetes Care. 2010;33(Suppl 1):S629. 85. Sabanayagam C, Khoo EY, Lye WK, et al. Diagnosis
70. Federation WHOaID. Definition and diagnosis of of diabetes mellitus using HbA1c in Asians: rela-
diabetes mellitus and intermediate hyperglycemia: tionship between HbA1c and retinopathy in a mul-
report of a WHO/IDF consultation. World Health tiethnic Asian population. J Clin Endocrinol Metab.
Organization Geneva; 2006. 2015;100:68996.
71. China Diabetes Society. China Guideline for Type 2 86. Qiao Q, Nakagami T, Tuomilehto J, et al. Comparison
diabetes. 2010 edition (in Chinese) Beijing: Peking of the fasting and the 2-h glucose criteria for diabetes
University Medical Press; 2011, p. 5. in different Asian cohorts. Diabetologia. 2000;43:
72. Philippines United for Diabetes Philippine practice 14705.
guidelines on diagnosis and management of diabetes 87. Ko GT, Chan JC, Yeung VT, et al. Combined use of
mellitus. Compendium of Philippine Medicine, 16th a fasting plasma glucose concentration and HbA1c
edition 2014. or fructosamine predicts the likelihood of having
73. Seino Y, Nanjo K, Tajima N, et al. Report of the diabetes in high-risk subjects. Diabetes Care. 1998;
committee on the classification and diagnostic crite- 21:12215.
ria of diabetes mellitus. J Diabetes Investig. 2010;1: 88. Ko G, Chan J, Cockram C. Use of a paired value of
21228. fasting plasma glucose and glycated hemoglobin in
74. Ko SH, Kim SR, Kim DJ, et al. 2011 clinical prac- predicting the likelihood of diabetes in a community.
tice guidelines for type 2 diabetes in Korea. Diabetes Diabetes Care. 1999;22:1909.
Metab J. 2011;35:4316. 89. Narayan KMV, Gregg EW, Engelgau MM, et al.
75. International Expert Committee report on the role of Translation research for chronic diseases. The case
the A1C assay in the diagnosis of diabetes. Diabetes for diabetes. Diabetes Care. 2000;23:17948.
Care. 2009;32:132734. 90. Fisher EB, Chan JCN, Nan H, Sartorius N,
76. Buell C, Kermah D, Davidson MB. Utility of A1C Oldenburg B. Co-occurrence of diabetes and depres-
for diabetes screening in the 1999 2004 NHANES sion: conceptual considerations for an emerging
population. Diabetes Care. 2007;30:22335. global health challenge. J Affect Disord. 2012;
77. Colagiuri S. Glycated haemoglobin (HbA1c) for the 142(Suppl):S5666.
diagnosis of diabetes mellitus practical implica- 91. Kong AP, Yang X, Ko GT, et al. Effects of treat-
tions. Diabetes Res Clin Pract. 2011;93:3123. ment targets on subsequent cardiovascular events in
78. Little RR, Rohlfing CL, Wiedmeyer HM, Myers GL, Chinese patients with type 2 diabetes. Diabetes Care.
Sacks DB, Goldstein DE. The national glycohemo- 2007;30:9539.
5 Diabetes in China and the Western Pacific Region 81

92. Leung WYS, So WY, Tong PCY, Chan NN, Chan 107. Inzucchi SE, Sherwin RS. The prevention of type
JCN. Effects of structured care by a pharmacist- 2 diabetes mellitus. Endocrinol Metab Clin N Am.
diabetes specialist team in patients with type 2 diabetic 2005;34:199219, viii.
nephropathy. Am J Med. 2005;118:1414.e421e427. 108. Brown N, Critchley J, Bogowicz P, Mayige M,
93. Chan JCN, Gagliardino JJ, Baik SH, et al. Multi- Unwin N. Risk scores based on self-reported or
faceted determinants for achieving glycaemic con- available clinical data to detect undiagnosed Type
trol: the International Diabetes Management Practice 2 diabetes: a systematic review. Diabetes Res Clin
Study (IDMPS). Diabetes Care. 2009;32:22733. Pract. 2012;98:36985.
94. Chuang LM, Soegondo S, Soewondo P, et al. 109. CDC Diabetes Cost-effectiveness Study Group. The
Comparisons of the outcomes on control, type of cost-effectiveness of screening for Type 2 diabetes.
management and complications status in early onset J Am Med Assoc. 1998;280:175763.
and late onset type 2 diabetes in Asia. Diabetes Res 110. Chan JC, So WY, Ma RCW, Tong P, Wong R,
Clin Pract. 2006;71:14655. Yang XL. The complexity of both vascular and
95. David AM, Rubio JM, Luces PS, Zabala RV, Roberto non-vascular complications of diabetes: the Hong
JP. Getting the patients perspective: a survey of diabe- Kong Diabetes Registry. Curr Cardiovasc Risk Rep.
tes services on Guam. Hawaii Med J. 2010;69:459. 2011;5:2309.
96. Wagner EH, Groves T. Care for chronic diseases. Br 111. Li JK, Ng MC, So WY, et al. Phenotypic and genetic
Med J. 2002;325:9134. clustering of diabetes and metabolic syndrome in
97. Marshall M, Pronovost P, Dixon-Woods Chinese families with type 2 diabetes mellitus.
M. Promotion of improvement as a science. Lancet. Diabetes Metab Res Rev. 2006;22:4652.
2013;381:41921. 112. Gong Q, Gregg EW, Wang J, et al. Long-term effects
98. Stock S, Pitcavage JM, Simic D, et al. Chronic care of a randomised trial of a 6-year lifestyle intervention
model strategies in the United States and Germany in impaired glucose tolerance on diabetes-related
deliver patient-centered, high-quality diabetes care. microvascular complications: the China Da Qing
Health Aff. 2014;33:15408. Diabetes Prevention Outcome Study. Diabetologia.
99. Tricco AC, Ivers NM, Grimshaw JM, et al. 2011;54:3007.
Effectiveness of quality improvement strategies on 113. Li G, Zhang P, Wang J, et al. Cardiovascular mortal-
the management of diabetes: a systematic review ity, all-cause mortality, and diabetes incidence after
and meta-analysis. Lancet. 2012;379:225261. lifestyle intervention for people with impaired glu-
100. Welin L, Wilhelmsen L, Bjornberg A, Oden A. Aspirin cose tolerance in the Da Qing Diabetes Prevention
increases mortality in diabetic patients without car- Study: a 23-year follow-up study. Lancet Diabetes
diovascular disease: a Swedish record linkage study. Endocrinol. 2014;2:47480.
Pharmacoepidemiol Drug Saf. 2009;18:11439. 114. Laatikainen T, Dunbar JA, Chapman A, et al.
101. Western Pacific Declaration on Diabetes (WPDD) Prevention of type 2 diabetes by lifestyle interven-
Steering Committee, On behalf of International tion in an Australian primary health care setting:
Diabetes Federation Western Pacific Region, World greater green triangle (GGT) diabetes prevention
Health Organization Western Pacific Region, project. BMC Public Health. 2007;7:249.
Secretariat of Pacific Community. Plan of Action 115. Sukala WR, Page R, Cheema BS. Targeting the type
(20062010) for the Western Pacific Declaration on 2 diabetes epidemic in Polynesia: historical perspec-
Diabetes: from Evidence to Action: WHO-WPRO tive and rationale for exercise intervention trials.
Publication Office. 2007. Ethn Dis. 2012;22:1238.
102. Wang HH, Wang JJ, Wong SY, Wong MC, Mercer 116. World Health Organization. Updated revised draft
SW, Griffiths SM. The development of urban com- global action plan for the prevention and control of
munity health centres for strengthening primary care noncommunicable diseases 20132020. 2013.
in China: a systematic literature review. Br Med 117. Tutino GE, Tam WH, Yang X, Chan JC, Lao
Bull. 2015;116:13954. TT, Ma RC. Diabetes and pregnancy: perspec-
103. Chen Z. Launch of the health-care reform plan in tives from Asia. Diabet Med J Br Diabet Assoc.
China. Lancet. 2009;373:13224. 2014;31:30218.
104. Willi C, Bodenmann P, Ghali WA, Faris PD, Cornuz 118. Urakami T, Suzuki J, Mugishima H, et al. Screening
J. Active smoking and the risk of type 2 diabetes: and treatment of childhood type 1 and type 2 diabe-
a systematic review and meta-analysis. JAMA J Am tes mellitus in Japan. Pediatr Endocrinol Rev PER.
Med Assoc. 2007;298:265464. 2012;10 Suppl 1:5161.
105. Pan A, Wang Y, Talaei M, Hu FB. Relation of 119. Wei J, Sung F, Lin C, Lin R, Chiang C, Chuang
smoking with total mortality and cardiovascular L. National surveillance for type 2 diabetes melli-
events among patients with diabetes mellitus: a tus in Taiwanese children. JAMA J Am Med Assoc.
meta-analysis and systematic review. Circulation. 2003;290:134550.
2015;132:1795804. 120. Toh C, Cutter J, Chew S. School based intervention
106. Mackay J, Ritthiphakdee B, Reddy KS. Tobacco has reduced obesity in Singapore. Br Med J. 2002;
control in Asia. Lancet. 2013;381:15817. 324:427.
82 J.C.N. Chan et al.

121. Chan JC, Deerochanawong C, Shera AS, et al. Role 135. Green JB, Bethel MA, Armstrong PW, et al. Effect
of metformin in the initiation of pharmacotherapy of sitagliptin on cardiovascular outcomes in type 2
for type 2 diabetes: an Asian-Pacific perspective. diabetes. N Engl J Med. 2015;373:23242.
Diabetes Res Clin Pract. 2007;75:25566. 136. Li Y, Xu W, Liao Z, et al. Induction of long-term
122. Nagi DK, Yudkin JS. Effects of metformin on insu- glycemic control in newly diagnosed type 2 diabetic
lin resistance, risk factors for cardiovascular disease, patients is associated with improvement of beta-cell
and plasminogen activator inhibitor in NIDDM sub- function. Diabetes Care. 2004;27:2597602.
jects. A study of two ethnic groups. Diabetes Care. 137. Weng J, Li Y, Xu W, et al. Effect of intensive insulin
1993;16:6219. therapy on beta-cell function and glycaemic control
123. Global guideline for type 2 diabetes. Diabetes Res in patients with newly diagnosed type 2 diabetes: a
Clin Pract. 2014;104:152. multicentre randomised parallel-group trial. Lancet.
124. Yang W, Weng J. Early therapy for type 2 diabe- 2008;371:175360.
tes in China. Lancet Diabetes Endocrinol. 2014;2: 138. Pozzilli P, Leslie RD, Chan J, et al. The A1C and
9921002. ABCD of glycaemia management in type 2 diabe-
125. Chan JC, Chan KW, Ho LL, et al. An Asian multi- tes: a physicians personalized approach. Diabetes
center clinical trial to assess the efficacy and toler- Metab Res Rev. 2010;26:23944.
ability of acarbose compared with placebo in type 2 139. Chan JC, Ozaki R, Luk A, et al. Delivery of inte-
diabetic patients previously treated with diet. Asian grated diabetes care using logistics and information
Acarbose Study Group. Diabetes Care. 1998;21: technology the Joint Asia Diabetes Evaluation
105861. (JADE) program. Diabetes Res Clin Pract. 2014;106
126. Yang W, Liu J, Shan Z, et al. Acarbose compared Suppl 2:S295304.
with metformin as initial therapy in patients with 140. Chan JC, Sui Y, Oldenburg B, et al. Effects of
newly diagnosed type 2 diabetes: an open-label, telephone-based peer support in patients with type
non-inferiority randomised trial. Lancet Diabetes 2 diabetes mellitus receiving integrated care: a ran-
Endocrinol. 2014;2:4655. domized clinical trial. JAMA Intern Med. 2014;174:
127. Chiasson JL, Josse RG, Gomis R, Hanefeld M, 97281.
Karasik A, Laakso M. Acarbose for prevention of 141. Kong AP, Yang X, Luk A, et al. Severe hypoglyce-
type 2 diabetes mellitus: the STOP-NIDDM ran- mia identifies vulnerable patients with type 2 diabe-
domised trial. Lancet. 2002;359:20727. tes at risk for premature death and all-site cancer:
128. Kawamori R, Tajima N, Iwamoto Y, Kashiwagi A, the Hong Kong diabetes registry. Diabetes Care.
Shimamoto K, Kaku K. Voglibose for prevention of 2014;37:102431.
type 2 diabetes mellitus: a randomised, double-blind 142. Ma RC, So WY, Tam CH, et al. Genetic variants for
trial in Japanese individuals with impaired glucose type 2 diabetes and new-onset cancer in Chinese
tolerance. Lancet. 2009;373:160714. with type 2 diabetes. Diabetes Res Clin Pract. 2014;
129. Chang CH, Chang YC, Lin JW, Chen ST, Chuang 103:32837.
LM, Lai MS. Cardiovascular risk associated with 143. Kong AP, Yang X, So WY, et al. Additive effects
acarbose versus metformin as the first-line treat- of blood glucose lowering drugs, statins and renin-
ment in patients with type 2 diabetes: a nation- angiotensin system blockers on all-site cancer
wide cohort study. J Clin Endocrinol Metab. risk in patients with type 2 diabetes. BMC Med.
2015;100:11219. 2014;12:76.
130. Holman RR, Bethel MA, Chan JC, et al. Rationale for 144. Jiao F, Fung CS, Wan YF, et al. Long-term effects
and design of the Acarbose Cardiovascular Evaluation of the multidisciplinary risk assessment and man-
(ACE) trial. Am Heart J. 2014;168:239.e2. agement program for patients with diabetes melli-
131. Cho YM. Incretin physiology and pathophysiol- tus (RAMP-DM): a population-based cohort study.
ogy from an Asian perspective. J Diabetes Investig. Cardiovasc Diabetol. 2015;14:105.
2015;6:495507. 145. Fung CS, Wan EY, Jiao F, Lam CL. Five-year change
132. Kim YG, Hahn S, Oh TJ, Kwak SH, Park KS, Cho of clinical and complications profile of diabetic
YM. Differences in the glucose-lowering efficacy patients under primary care: a population-based lon-
of dipeptidyl peptidase-4 inhibitors between Asians gitudinal study on 127,977 diabetic patients. Diabetol
and non-Asians: a systematic review and meta- Metab Syndr. 2015;7:79.
analysis. Diabetologia. 2013;56:696708. 146. Wong CK, Wong WC, Wan YF, et al. Patient
133. Kim YG, Hahn S, Oh TJ, Park KS, Cho Empowerment Programme in primary care reduced
YM. Differences in the HbA1c-lowering efficacy of all-cause mortality and cardiovascular diseases in
glucagon-like peptide-1 analogues between Asians patients with type 2 diabetes mellitus: a population-
and non-Asians: a systematic review and meta- based propensity-matched cohort study. Diabetes
analysis. Diabetes Obes Metab. 2014;16:9009. Obes Metabol. 2015;17:12835.
134. Scirica BM, Bhatt DL, Braunwald E, et al. 147. Raz I, Riddle MC, Rosenstock J, et al. Personalized
Saxagliptin and cardiovascular outcomes in patients management of hyperglycemia in type 2 diabetes:
with type 2 diabetes mellitus. N Engl J Med. reflections from a Diabetes Care Editors Expert
2013;369:131726. Forum. Diabetes Care. 2013;36:177988.
5 Diabetes in China and the Western Pacific Region 83

148. Institute of Medicine Crossing the quality chasm: a 2015. Accessed 9th Feb 2016, at
new health system for the 21st century. 2001. http:// centres-education. 153. WHO. WHO definition of health. Preamble to the
149. Yin J, Wong R, Au S, et al. Effects of providing Constitution of the World Health Organization as
peer support on diabetes management in people adopted by the International Health Conference,
with type 2 diabetes. Ann Fam Med. 2015;13 Suppl New York, 1922 June 1946, and entered into force
1:S429. on 7 April 1948. 1948.
150. United Nation General Assembly. Political dec- 154. Western Pacific Declaration on Diabetes Steering
laration of the High-level Meeting of the General Committee. Plan of action (20062010) for the
Assembly on the Prevention and Control of Non- Western Pacific declaration on diabetes: from evi-
communicable Diseases A/66/L.12011. dence to action. Manila: World Health Organization
151. Holmes D. Margaret Chan: committed to universal Western Pacific Regional Office; 2008. Available at
health coverage. Lancet. 2012;380:879.
152. International Diabetes Federation Reference Centre
of Education
Diabetes in India and Southeast
Asia 6
Shashank R. Joshi and S.R. Aravind

Epidemiological Trends Moreover, Asians have a strong ethnic and

and Emerging Disease Burden genetic predisposition for diabetes and have
lower thresholds for the environmental risk fac-
Diabetes is a multisystem disorder associated tors. There are 387 million people with diabetes
with complications, and the prevalence of which in the world with 78.3 million people in the SEA
is increasing globally. Diabetes imposes region which is expected to rise to 131 million
immense public health burden with unaccept- by the year 2040.
ably high burdens on individuals, their families, The last three decades have witnessed an epi-
and national economies. As the urbanrural demic rise in the number of people with diabetes,
divide narrows consistently, it adversely affects especially type 2 diabetes, and particularly in
the lifestyle of populations. The rapid emerging developing countries, where more than 80 % of
economies of Southeast Asia (SEA) are a victim the people with diabetes live. The recent land-
to the epidemiological transition which results in mark study for the pooled analysis across 751
the shifting of the disease burden from the com- studies with 4.4 million adults from 200 coun-
municable to the non-communicable diseases. tries published in Lancet reflects that between
1980 and 2014 the number of adults with diabe-
tes in the world increased fourfold from 108 mil-
lion to 422 million. The increase has particularly
Joshi Clinic, Lilavati & Bhatia Hospital, been sharp in low- and middle-income countries.
Mumbai, India In 2014, 50 % of adults with diabetes lived in five
Endocrine Society of India, New Delhi, India
Association of Physicians of India (API) (2014-15),
Columbia Asia Hospital,
Mumbai, India
Yeshwanthpur, Bangalore, India
Indian Academy of Diabetes, Mumbai, India
Post Graduate Studies in Diabetology (RGUHS &
RSSDI (Research Society for the Study of Diabetes RSSDI), Bangalore, India
in India 2011), Mumbai, India
PESIMSR, Kuppam, India
AIAARO (All India Association of Advancement for
RSSDI (Research Society for Study of Diabetes in
Research in Obesity), Mumbai, India
India), Bangalore, India
DiabetesIndia, Hyderabad, India
Diacon Hospital, Bangalore, India KRSSDI, Bangalore, India

Springer International Publishing Switzerland 2017 85

S. Dagogo-Jack (ed.), Diabetes Mellitus in Developing Countries and Underserved Communities,
DOI 10.1007/978-3-319-41559-8_6
86 S.R. Joshi and S.R. Aravind

countries: China, India, the USA, Brazil and the region during 2015. India is home to the
Indonesia. The prevalence of diabetes in adults second largest number of children with type 1
more than doubled for men in India and China diabetes in the world (70,200), after the USA,
(3.79.1 % in India, 3.59.9 % in China) but and accounts for the majority of the children
increased by 80 % amongst women in India (4.6 with type 1 diabetes in the region. More than
8.3 %) but only 50 % in women in China half (53.2 %) of these deaths occurred in people
(57.6 %). The total number of adults with diabe- under 60 years of age.
tes in India increased from 11.9 million in 1980 With an estimated 69.2 million people suffer-
to 64.5 million in 2014. In China, the increase ing from the condition, the largest in any country
was from 20.4 million in 1980 to 102.9 million in in the world, diabetes has become a major health-
2014. While India contributed 15.3 % of the care problem in India (refer to Figs. 6.2 and 6.3).
global share of adults with diabetes in 2014, it Recent epidemiological studies from India point
was 24.4 % for China. Across the region, approx- to the great burden due to diabetes and its micro-
imately 72 million people have diabetes close and macrovascular complications. This is primar-
to one fifth of all adults with diabetes in the ily because the status of diabetes control in India
world. Overall, the rise of type 2 diabetes in is far from ideal. Based on the available data, the
South Asia is estimated to be more than 150 % mean glycated haemoglobin levels are around 9 %
between 2000 and 2035. which is at least 2 % higher than the goal currently
suggested by international bodies. A balanced
approach to improve awareness about diabetes
Economic and the Societal Impact and its control both amongst patients and the
of the Complications medical fraternity is an urgent need of the hour in
Diabetes is increasingly affecting individu-
als in the region in their most productive years
(refer to Fig. 6.1). This will pose a challenge to Factors Contributing to the Rapid
governments working to improve the economic Increase in Prevalence of Diabetes
situation in their countries. The scenario poses in Asia
huge social and economic problems to most
nations in the region and could impede national Although ageing, urbanisation and associated
and, indeed, global development. More than half lifestyle changes are the major determinants for
of the deaths due to diabetes occur in people the rapid increase, an adverse intrauterine
under 60 years of age and one quarter in people environment and the resulting epigenetic changes
under 50 years of age. India is the largest con- could also contribute in many developing coun-
tributor to regional mortality, with 1.1 million tries. More action is required to understand the
deaths attributable to diabetes in 2015. Despite drivers of the epidemic to provide a rationale for
the huge number of people with diabetes in the prevention strategies to address the rising global
Southeast Asia Region, health-care spending on public health tsunami.
diabetes was estimated to be only USD 6 billion,
accounting for less than 1 % of the global total,
with India estimated to have spent the largest Urbanisation and Socioeconomic
proportion. The health-care spending appears to Transition
be low as government spending on healthcare
is privatised less as predominant healthcare in Diabetes burden in India is contributed by vari-
India. There are an estimated 81,400 children ous factors. Genetic predisposition combined
under the age of 15 living with type 1 diabetes with lifestyle changes, associated with urbanisa-
in the Southeast Asia Region. Approximately tion and globalisation, contributes to this rapid
13,100 children developed type 1 diabetes in rise of diabetes in India. The highest rates of
6 Diabetes in India and Southeast Asia 87

Prevalence of diabetes in adults by age, 2014





Source: IDF Diabetes Atlas, Update 2014

Prevalence (%)

2024 2529 3034 3539 4044 4549 5054 5559 6064 6569 7074 7579
Age (years)

Fig. 6.1 Age-wise prevalence of diabetes

1980 2014
Rank Country Millions of adults with diabetes Rank Country Millions of adults with diabetes
(% of global diabetes) (% of global diabetes)
1 China 20-4 (18-9) 1 China 102-9 (24-4)
2 India 11-9 (11-0) 2 India 64-5 (15-3)
3 USA 8-1 (7-5) 3 USA 22-4 (5-3)
4 Russia 7-1 (6-6) 4 Brazil 11-7 (2-8)
5 Japan 4-7 (4-4) 5 Indonesia 11-7 (2-8)
6 Germany 3-4 (3-2) 6 Pakistan 11-0 (2-6)
7 Brazil 2-7 (2-5) 7 Japan 10-8 (2-6)
8 Ukraine 2-4 (2-2) 8 Russia 10-7 (2-5)
9 Italy 2-4 (2-2) 9 Egypt 8-6 (2-0)
10 UK 2-3 (2-1) 10 Mexico 8-6 (2-0)

12 Indonesia 2-1 (1-9)

13 Pakistan 1-7 (1-6)
14 Germany 5-1 (1-2)
15 Mexico 1-7 (1-6)
16 Italy 4-3 (1-0)
17 Egypt 1-5 (1-4)

19 UK 3-8 (0-9)

24 Ukraine 3-4 (0-8)

Fig. 6.2 Ten countries with the largest no. of diabetes in 1980 and 2014 [3]

urbanisation have been in Singapore, Korea, intermediate rates (30 %) and Bangladesh and Sri
Malaysia, the Philippines and Indonesia (50 %). Lanka have slow rates of urbanisation. The
China, Pakistan, India and Thailand have increase in urban population and ageing are the
88 S.R. Joshi and S.R. Aravind

2015 more prevalence at the age of 6069 years,

whereas in the Chinese population, it peaks at

Number of people
with diabetes
7989 years. Indians also have a higher preva-
1 China 109.6 million [99.6133.4]
lence of impaired glucose tolerance at a younger
2 India 69.2 million [56.284.8]
age than the Chinese population. The findings
3 United States of America 29.3 million [27.630.9] from Pakistan and Sri Lanka also show similar
4 Brazil 14.3 million [12.915.8] results.
5 Russian Federation 12.1 million [6.217.0]

6 Mexico 11.5 million [6.213.7]

Anthropometry: Thin-Fat Phenotype
7 Indonesia 10.0 million [8.710.9]

8 Egypt 7.8 million [3.89.0] It is observed that amongst Asians, diabetes

9 Japen 7.2 million [6.19.6] occurs at lower body mass index (BMI) levels
10 Bangladesh 7.1 million [5.312.0]
than in Western populations, and small incre-
ments in weight trigger glucose intolerance in
Fig. 6.3 Prevalence of diabetes in different countries susceptible subjects. Especially Asian Indians
(age 2079 years) [1] have higher odds of developing diabetes, despite
having a significantly lower BMI than the white
main determinants of the global rise in the preva- population.
lence of diabetes. Urbanisation and internal rural Several studies in Asian populations, particu-
to urban migration result in several adverse larly in Asian Indians, have highlighted the met-
impacts: physical activity decreases, diet habits abolically obese phenotype amongst normal
shift towards high-energy foods and body mass weight individuals. This phenotype, character-
index (BMI) and upper body adiposity increase ised by greater abdominal obesity despite a nor-
considerably. The Indian Council of Medical mal BMI, less muscle mass, higher percentage of
Research (ICMR) study done in the 1970s body fat and increased propensity for insulin
reported a prevalence of 2.3 % in urban areas resistance compared with the Western popula-
which has risen to 1219 % in the 2000s. tion, renders higher susceptibility for diabetes in
Correspondingly, in rural areas, prevalence rates Asian populations.
have increased from around 1 to 410 % and even The association of BMI and diabetes is well
13.2 % in one study. Thus it is clear that both in established and is usually modified by ethnicity.
urban and rural India, prevalence rates of diabe- Ethnicity factors that contribute are genetic con-
tes are rising rapidly with a rough urbanrural stitution, lifestyle, living environment and anthro-
divide of 2:1 or 3:1 being maintained through the pometric characteristics. Body composition
last two to three decades with the exception of related to fat distribution is a stronger determinant
Kerala where rural prevalence rates have caught of the metabolic milieu than BMI. The diabetes
up with or even overtaken urban prevalence rates. epidemiology, collaborative analysis of diabetes
The postulates from the ICMRINDIAB study criteria in Europe/in Asia study group noted that
predicted for the burden of diabetes which pro- the overall effect of age on the prevalence of dia-
jected that in 2011, India would have 62.4 million betes differed considerably between the ethnic
people with diabetes and 77.2 million people groups, even in the subjects with the same
with prediabetes. BMI. Asian populations are prone to have more
intra-abdominal fat accumulation and low muscle
mass. Asian Indians, in particular, have the above
Age abnormalities which account for the high preva-
lence of insulin resistance and diabetes at low lev-
As compared to Western populations, Asian els of BMI. The risk of diabetes increases
Indians develop diabetes at a younger age with progressively from a BMI of 23 kg/m2 amongst
6 Diabetes in India and Southeast Asia 89

Indians. BMI in of 23 kg/m2 is also considered advantageous in certain populations during evo-
overweight for most Asian populations. Asian lutionary selection by repeated famine and feast
Indians have small body size which has been cycles. However, these genes have rendered them
named as thin-fat Indian. Asian Indians have highly predisposed to obesity and diabetes during
thinner limbs, which are suggestive of smaller the modern era of continuous feasting. On the
muscle mass. However, despite their thinness, other hand, the thrift phenotype hypothesis pos-
they are centrally obese, with higher waisthip tulates that intrauterine malnutrition leads to met-
ratio (WHR) and higher subscapulartriceps skin- abolic and structural changes in the beta cells that
fold ratio than their British counterparts. Many are beneficial for early survival, but increases the
studies show that Asian Indians have more body risk of T2D and other chronic disorders in
fat for any given BMI compared with Caucasians adulthood.
and black Africans. Indians also have higher lev-
els of central obesity (measured as waist circum-
ference [WC], WHR, visceral fat and posterior Screening
subcutaneous abdominal fat). This is reflected in
higher plasma nonesterified fatty acid (NEFA) The prevalence of the micro- and macrovascular
and triglyceride concentrations, hyperinsulinae- complications also influences the mortality rate
mia with fasting as well as post-glucose challenge due to diabetes. Unfortunately, more than 50 % of
states and higher insulin resistance. Thus, Asian individuals with diabetes in India remain undiag-
Indians have an unusual thin-fat body composi- nosed, and some may even present with macro-
tion associated with the insulin resistance syn- vascular disease (coronary artery disease and
drome, and this is the now popular thin-fat cerebrovascular disease or stroke and peripheral
Indian concept. vascular disease) and microvascular disease (reti-
nopathy, nephropathy and neuropathy) at the
time of diagnosis.
Smoking and Alcohol Use Data on various complications of diabetes
have also been published by several authors.
Smoking increases the risk of central obesity and However, till recently, most of such data were
insulin resistance and the risk of diabetes is hospital or clinic based and therefore subject to
shown to be higher by 45 % in smokers than referral bias. The Chennai Urban Rural
amongst nonsmokers. The increasing use of alco- Epidemiology Study (CURES) and the Chennai
hol in Asian countries, especially amongst the Urban Population Study (CUPS) provide the first
middle class and rural population, also increases population-based data from India on virtually all
the risk for diabetes and other metabolic diseases complications of diabetes. CURES was a
and deleterious health effects. population-based study involving 26,001 partici-
pants aged 20 years or above based on a represen-
tative population of Chennai. The overall
Genetic Susceptibility prevalence of diabetic retinopathy based on four-
field stereo colour retinal photography was
The genetic burden on Asian Indians makes the 17.6 %. The prevalence of overt nephropathy was
population more susceptible to diabetes. This risk 2.2 % while that of microalbuminuria was 26.9 %.
is further increased due to interaction with envi- Peripheral neuropathy based on biothesiometry
ronmental triggers. Exposure to a high fat diet was detected in 26.1 %.
and lower levels of physical activity trigger the In the CUPS study, coronary artery disease
geneenvironmental interaction. Both the thrifty was evident in 21.4 % of diabetic subjects, 14.9 %
genotype and thrifty phenotype hypotheses of subjects with impaired glucose tolerance and
appear to be the aetiology. The selective presence in 9.1 % of people with normal glucose tolerance.
of thrifty genotypes has been considered to be In the same study, peripheral vascular disease
90 S.R. Joshi and S.R. Aravind

was present in 6.3 % of diabetic subjects com- awareness levels increased with education,
pared to 2.7 % amongst nondiabetic subjects. only 42.6 % of postgraduates and profession-
Diabetic subjects also had increased subclinical als, which group included doctors and lawyers,
atherosclerosis as measured by intimal-medial knew that diabetes was preventable. The knowl-
thickness at every age point, compared to their edge of risk factors of diabetes was even lower
nondiabetic counterparts. Assuming that 40 mil- with only 11.9 % of the study subjects reporting
lion people in India have diabetes, this translates obesity and physical inactivity as risk factors for
to at least 7 million with retinopathy, 0.8 million diabetes. More alarming was the fact that even
with nephropathy, 10.4 million with neuropathy, amongst known diabetic subjects, only 40.6 %
8.5 million with CAD and 2.5 million with were aware that diabetes could lead to some
PVD. Thus, the burden due to diabetic complica- organ damage. There is another population-
tions is very high in India due to the sheer num- based study which was done to find out the levels
ber of people with diabetes. These figures are in and details regarding awareness on diabetes in
fact very conservative, and it is possible that in urban adult Indian population aged 20 years.
rural areas, the prevalence of complications is The knowledge regarding the causes of diabe-
much higher because of poorer control of diabe- tes, its prevention and the methods to improve
tes and lack of access to healthcare. the health was significantly low amongst the
Identifying accurate and low-cost screening general population. In the total study group,
methods is a necessary first step in assessing the 41 % were unaware of the health being affected
cost-effectiveness of screening to detect undiag- by diabetes, and only less than 30 % knew about
nosed diabetes. Indian Diabetes Risk Score the complications related to kidneys, eyes and
(IDRS) is more effective and significantly less nerves. Many persons with diabetes (46 %) felt
expensive for screening for undiagnosed T2DM it was a temporary phenomenon. Amongst the
compared to genotyping TCF7L2 SNPs, the diabetic subjects, 92.3 % had sought the help of
strongest genetic marker for T2DM currently a general practitioner to take treatment. Only a
available. Using IDRS screening prior to OGTT small proportion went to a specialist.
reduces costs while still detecting a substantial
portion of NDD individuals. A potential addi-
tional benefit of both the IDRS and genotyping is Current Status of Diabetes Control
their ability to identify individuals who currently in India
do not have diabetes but are at high risk of devel-
oping diabetes in the future. Thus an individual The next challenge in India is that the quality of
with an IDRS score of 60 at baseline was three diabetes care varies considerably depending upon
times more likely to develop diabetes in the the awareness levels, expertise available, atti-
future than low-risk subjects (IDRS <30). tudes and perceptions amongst diabetes care pro-
viders. An estimate based on sales of antidiabetic
pharmaceutical agents shows that on an average
Awareness of Diabetes in India only 1012 % of people with diabetes receive
modern pharmacological treatment in India. In
The awareness of diabetes is a cornerstone of 1998, the DiabcareAsia study was carried out to
the prevention of this disease. CURES reported investigate the relationship between diabetes
that nearly 25 % of the population was unaware control, management and late complications in a
of diabetes. Only around 40 % of the partici- subset of urban Indian diabetes population treated
pants felt that the prevalence of diabetes was at 26 tertiary diabetes care centres. A total of
increasing, and only 22.2 % of the popula- 2,269 patients participated in this study and it
tion and 41 % of known diabetic subjects felt was observed that approximately half of the
that diabetes could be prevented. Though the patients had poor control (HbA1c >2 % points
6 Diabetes in India and Southeast Asia 91

above upper limit of normal), and the mean diet-related non-communicable diseases like
HbA1c was significantly higher (8.9 2.1 %) T2DM, cardiovascular disease, etc. predomi-
than the levels recommended by the American nantly in urban, but also in rural areas. The
Diabetes Association and the ICMR guidelines in dietary recommendations should be individual-
India. Over 54 % of patients had diabetes-related ised according to the persons ethnicity, cultural
complications. The mean HbA1c levels and fre- and family background and personal prefer-
quency of complications were higher in patients ences and associated comorbid conditions. It
with longer diabetes duration. This study also should be flexible in a variety and preparation
showed that 4 % of patients were on diet therapy, of food choices and timing of meals according
53.9 % were receiving oral antidiabetic agents to the persons daily routine. Both the National
(OHAs), 22 % were receiving insulin and 19.8 % Institute of Nutrition and expert group [2] have
were receiving a combination of insulin and developed some broad Indian guidelines which
OHAs. This study concluded that with increasing recommend reduction intake of carbohydrate,
duration of diabetes, glycaemic control deterio- higher intake of fibre, lower intake of saturated
rates leading to late complications. It also con- fat, optimal ratio of essential fatty acids, slightly
firmed that diabetes care in India leaves much to higher protein intake, lower intake of salt and
be desired and suggested the need for efforts to restricted intake of sugar.
increase awareness amongst health professionals The role of regular physical activity is well
to improve diabetes care in India. established in the management in persons with
type 2 diabetes. A careful assessment of an indi-
vidual should be made by the physician while
Non-pharmacologic Approach incorporating an exercise programme in the man-
for the Management of Diabetes agement. Exercise programme should be indi-
vidualised according to the individual capacity
Lifestyle modifications are the cornerstone of and disabilities. The person with diabetes must
management of diabetes mellitus and include wear appropriate footwear.
the prescription of a healthy diet, regular exer-
cise, the management of stress and avoidance of
tobacco. The aims of dietary management are Clinical Inertia in Diabetes: Failure
to achieve and maintain ideal body weight, eug- to Achieve Tight Control
lycaemia and desirable lipid profile, to prevent
and postpone complications related to diabetes Failure of initiation of or intensification of ther-
and to provide optimal nutrition during preg- apy, when indicated, is termed clinical inertia.
nancy, lactation, growth, old age and associ- Though we have well-defined management
ated conditions, e.g. hypertension and catabolic goals, effective therapies and practice guide-
illnesses. Recently the published STARCH lines, there is often a failure to take appropriate
study shows that Indians consume high carbo- action despite recognition of the problem. This is
hydrate in their diet compared to the Western a common problem in the management of
population. The comparison of macronutrients patients with asymptomatic chronic illnesses.
(i.e. region-wise carbohydrate, fat and protein) The use of soft reasons to avoid intensification
revealed a similar pattern of dietary consump- of therapy and lack of education, training and
tion, that is, high carbohydrate and a lower practice organisation aimed at achieving thera-
range of fat and protein. This study neutralises peutic goals are the common reasons for clinical
the myth that only the south Indian population inertia. Clinical inertia in achieving glycaemic
consumes high carbohydrate in their diet (rice, targets in Indian diabetic subjects could be
idli, etc.). Dietary transition and a sedentary expected to be even more than in the West, where
lifestyle have led to an increase in obesity and it has been reported that 65 % of the patients
92 S.R. Joshi and S.R. Aravind

diagnosed with diabetes, only 73 % are pre- goals with the potential for long periods of
scribed pharmacologic therapy and only 33 % of hyperglycaemia. Periods of hyperglycaemia
those treated achieve a haemoglobin A1C value long or even short can increase the risk of
of less than 7 % by the ADA goal. This may be micro- and macrovascular complications. The
due to the low rates of awareness of diabetes and current understanding of the complex patho-
its complications in India resulting in poor gly- physiology of the disease and the progressive
caemic control seen in Indians with diabetes. deterioration in glycaemic control over time
Moreover, other factors like poverty, lack of supports the philosophy of earlier intervention
accessibility to health services and inadequate with a more comprehensive initial therapy. The
follow-up are additional factors in developing major classes of antidiabetic agents that may
countries like India. be combined with metformin include sulpho-
Consequently, insulin is delayed until it is nylurea (SU), thiazolidinedione (TZD), dipep-
absolutely necessary. Most patients are initiated tidyl peptidase-4 inhibitor (DPP4-i), insulin
on insulin after a course of multiple oral antidia- and glucagon-like peptide-1 (GLP-1) recep-
betic drugs. Insulin therapy is initiated only when tor agonist. Few studies have investigated the
the HbA1c levels had deteriorated further to effect of metformin-based early combination
around 9 %. Physicians often delay insulin ther- therapy. There are several different types of
apy worrying that the daily injections, modifica- insulin available, but as a minimum, regular
tion of lifestyle due to insulin and dependence on quick-acting human insulin and longer-acting
insulin for life and that patients may feel that NPH insulin should be available to everyone in
insulin therapy indicates the last stage of diabe- all parts of the world.
tes. However, patients who had moved on to insu- In India, which is a resource-limited coun-
lin seemed to have a more positive approach try, all therapies are available and it is a pre-
towards his/her treatment due to the improve- dominantly non-reimbursed market. Usually
ment in quality of life and better control despite sulphonylureas, metformin, alpha glucosidase
the issues outlined above. inhibitors and glitazone form the cornerstone of
therapy with insulin. However, recently gliptins
including the low-cost one as well as SGLT2
Pharmacologic Therapy inhibitors are also available. Biosimilar insulin is
in Diabetes: Is It Different? also available but not popular and premixed insu-
lin is still used widely. Cost and dose play a role
A proactive approach to treating type 2 diabetes in resource-limited environment. Indian usually
is recommended: therapy should be individual- requires lower doses and is more insulin resistant.
ised with early consideration of combination
therapy and ongoing reinforcement of lifestyle
modification messages. Indeed, the conser- Translating Primary Prevention
vative stepwise approach to type 2 diabetes of Diabetes
management involves lifestyle modification,
followed by treatment with a single oral anti- The Indian Diabetes Prevention Programme
diabetic agent, often up-titrated to maximal (IDPP) has been a unique prospective study
recommended doses before combination ther- which has provided several pathways and strat-
apy is introduced. Very often there is a delay egies for the prevention of diabetes in India
between stepping up from monotherapy (e.g. including the importance of the lifestyle modifi-
metformin alone) to combination therapy (e.g. cation and metformin which independently could
metformin plus other OADs, often sulpho- reduce the incidence of diabetes in Asian Indians
nylurea), and this can result in unacceptable with impaired glucose tolerance. Also, these have
delays in achieving and maintaining glycaemic been proposed as the cost-effective benchmarks
6 Diabetes in India and Southeast Asia 93

amongst high-risk individuals with high degree incident of prediabetes. The mechanistic insights
of insulin resistance and may be useful in other now ascribe these benefits through improvement
developing countries as well. It is important to in insulin sensitivity and beta-cell preservation.
control the persistent IGT as it is demonstrated Prospective, parallel-group, randomised con-
to add to the higher incidence of diabetes with trolled trial across close to 9,000 subjects have
other risk factors for diabetes, such as high BMI, demonstrated that mobile phone messaging is
waist circumference and body fat percentage. In an effective and acceptable method to deliver
a recent collaborative work across South Asia, advice and support towards lifestyle modification
Latin America and South Africa to compare the to prevent type 2 diabetes in men at high risk.
prevalence, awareness, treatment and control Evidence from the DPP, and other prevention tri-
of diabetes and assess the relationship between als conducted in patients with prediabetes, shows
diabetes and prediabetes with known cardio- that appropriate lifestyle modification includ-
vascular and metabolic risk factors, it has been ing physical activity could lead to risk reduc-
demonstrated that propensity for South Asians tion in the incidence of T2DM by almost 58 %.
to develop diabetes and prediabetes at a younger Studies have shown that resistance and aerobic
age and lower body mass index compared with exercise is effective in improving metabolic pro-
individuals from other low- and middle-income file of adults with T2DM. Previous research has
countries. Therefore, it is important that the long- reported improved insulin sensitivity/resistance
term impact and the complications are prevented, and reductions in hyperglycaemia-related medi-
and the health systems and policy makers must cations as a result of exercise training. In par-
make concerted efforts to improve diabetes pre- ticular, supervised resistance training (max. ten
vention and detection in the targeted population. repetitions for >3 days per week) has been shown
Ramachandran A et al. have suggested that it is to lead to significant improvement in insulin
important to develop precise predictors for inci- sensitivity and values of glycosylated haemoglo-
dent diabetes amongst Asian men. The analysis bin, lipid profile and truncal and peripheral sub-
of the data from the combined cohorts of the cutaneous adipose tissue in Asian Indians with
Indian Diabetes Prevention Programmes 1 and T2DM. It has been reported that children and
2 demonstrates that the baseline HbA (1c) was adolescents with type 1 diabetes should complete
highly predictive of future diabetes in Asian a minimum of 3060 min of moderate-intensity
Indian subjects with impaired glucose tolerance physical activity daily. Additional physical activ-
and nearly 60 % of the incidence occurred with ity beyond 60 min/day would be helpful in main-
values 6.0. Diagnostic sensitivity of 6.5 % taining glycaemic profile for T2DM patients. The
for new diabetes was only 51 % using the oral practice of yoga is a traditional Indian practice
glucose tolerance test as the standard for com- that helps therapeutically and promotes physical
parison. The combination of gamma-glutamyl- and mental health. Yoga-based lifestyle modifi-
transferase (GGT) and fasting plasma glucose cation programme helps in the reduction of blood
(FPG) offers a simple and sensitive tool to iden- glucose, HbA1c, triglycerides, total cholesterol
tify subjects at high risk of developing diabetes. and VLDL. Mindfulness eating and yoga exer-
Similarly, several other markers including adi- cise had health benefits on glycaemic control
ponectin, IL-6, retinol-binding protein 4, and in pregnant women with GDM in some studies.
hypertriglyceridaemic waist phenotype have Yogic exercises have enhanced the antioxidant
been proposed to independently associate with defence mechanism in diabetics by reducing
incident diabetes. Prospective, intervention stud- oxidative stress. Unless drastic steps are taken
ies have demonstrated that increased compliance through national prevention programmes to curb
to lifestyle goals especially with the modifica- the escalating trends in all of the countries, the
tion of the diet habits, independent of the physi- social, economic and health-care challenges are
cal activity, could result in the decrease in the likely to be insurmountable.
94 S.R. Joshi and S.R. Aravind

Organising and Conducting be further streamlined by an individualised

Diabetes Research in the Region approach (Fig. 6.4).

RSSDI (Research Society for the Study of

Diabetes in India) is the largest organisation of Future Directions: Unmet Needs,
diabetes health-care professionals and research- Unanswered Questions
ers in Asia, which was formed in 1972. Currently, and Unquestioned Answers
there are more than 5,500 life members from
across the country representing 29 Indian states There have been rapid epidemiological transitions
and Union territories. Every year, RSSDI organ- translating into a huge disease burden in diabetes
ises the national annual meeting, which not only in the SEA region. Prevention of diabetes through
provides a platform for its members to listen to consistent awareness about the disease in popula-
the leaders in the field of diabetes from within the tion would be a critical step forward which would
country as well as from abroad but also to interact have tremendous implications that halt the prog-
amongst themselves and exchange knowledge ress of disease. One of the global targets for non-
and ideas. Annual meetings of RSSDI have been communicable disease is to halt by 2025 the rise
a regular feature for more than four decades and in the prevalence of diabetes to its 2010 levels. A
are very well attended. RSSDI has a nationwide better understanding for the basis for the gene
presence through its 14 state chapters. All state environment interaction, in which beta-cell dys-
chapters carry on the work of RSSDI at the state function, typically on the background of insulin
and local level. In addition, these chapters carry resistance, is critical for the increase in glucose
out independent activities including CMEs for levels observed in impaired glucose metabolism
member physicians, local research grants and and for the development of the hyperglycaemia of
awareness programmes for public as well as dia- type 2 diabetes, would be explored to target effec-
betes patients. RSSDI regularly publishes a tive therapies. Prevention is of utmost importance,
newsletter, both in print and electronic format, but for the more than 420 million people currently
which serves as an important link between the living with diabetes, managing their disease must
national body and its membership to keep the remain the priority. The recent WHOs report rec-
members informed of various activities, research ommends a multidisciplinary approach with
grants and educational initiatives. The patient education, medication and consistent fol-
International Journal of Diabetes in Developing low-up. For primary prevention, the challenge lies
Countries (IJDDC) is the prestigious indexed in raising awareness, promoting health literacy
publication of RSSDI and is an important and identifying individuals at high risk of diabetes
resource of research work done in the field of dia- for early intervention. For secondary prevention,
betes in India. RSSDI funds research proposals poor access to care, clinical inertia and treatment
from Indian scientists interested in conducting non-adherence are major barriers in evidence-
research in the field of diabetes mellitus. For pro- based practice. Given the phenotypic heterogene-
viding research grants, RSSDI invites proposals ity of diabetes, and thus the pluralistic needs of
from Indian scientists interested in conducting those affected, clinical acumen to identify prob-
original research in the field of diabetes mellitus. lems and sufficient contact time to empower the
Furthermore, limited grants are also available for person to change behaviour and adhere to treat-
the students of medical colleges for smaller proj- ment are key to successful management.
ects. Recently, RSSDI has developed a simple Screening in young-onset diabetes in India for
user-friendly novel approach to decide the appro- CV risk factors and complications would be vital
priate antidiabetic agent to be used in type 2 dia- to curb the impact of the microvascular and mac-
betes through the therapeutic wheel. The best rovascular complications. This has been clearly
choices can be determined from the outer rings of demonstrated in the landmark CINDI and CINDI
the wheel (orange and red), and the choices can 2 trials published recently for the clinic-based
6 Diabetes in India and Southeast Asia 95

Fig. 6.4 The RSSDI therapeutic wheel

miological, experimental human and animal

survey amongst 4,600 patients for diabetes- studies as well as the data from several mega-
related complications and a retrospective cross- trials have convincingly proved the importance
sectional study of 1,500 patients with newly of tight metabolic control in arresting and pre-
detected young-onset diabetes, respectively. venting the progression of target organ dam-
age. In the last two decades, there is a better
Conclusions understanding of the pathophysiology of type 2
Considering the enormous burden due to dia- diabetes and availability of newer oral drugs
betes in India, it is important to realise the cost- for diabetes; newer insulin and improved deliv-
effective measures of diabetes care like early ery systems should translate to improve diabe-
screening, tight metabolic control, monitoring tes control. However, the survey described
of risk factors and assessing of the organ dam- above indicates the gaps between the guide-
age. The study done for economic analysis in lines and real-life practice. In view of this,
diabetes care in India has also shown that the appreciation and understanding of both patient
cost of providing routine care is only a fraction and physician barriers regarding proper moni-
of the overall cost and is perhaps still manage- toring and judicious use of therapeutic options
able. However, when this is not available or its including insulin therapy for optimising diabe-
quality is poor, the overall direct and indirect tes management should be encouraged in order
costs escalate with disastrous health and eco- to improve control of diabetes in India. Result-
nomic consequences to the individual, his fam- oriented organised programmes involving
ily and society particularly due to the onset of patient education, updating medical fraternity
the micro- and macrovascular complications of on various developments in the management of
diabetes. Published data from several epide- diabetes and providing them the opportunity to
96 S.R. Joshi and S.R. Aravind

use and analyse these newer treatment options 6. Mohan V, Alberti KGMM. Diabetes in the tropics.
in the form of observational studies are required In: Alberti KGMM, Zimmet P, Defronzo RA, et al.,
editors. International text book of diabetes mellitus.
to combat the diabetes epidemic currently 2nd ed. Chichester: Wiley; 1997. p. 17187.
threatening to affect the lives of millions of 7. Nakagami T, Qiao Q, Carstensen B, The DECODE
people in India. Coordination, patient educa- -DECODA Study Group, et al. Age, body mass
tion and ongoing support are important compo- index and type 2 diabetes-associations modified by
ethnicity. Diabetologia. 2003;46:106370.
nents of quality diabetes care, but most 8. Banerji MA, Faridi N, Atluri R, et al. Body composi-
health-care systems are created to provide tion, visceral fat, leptin, and insulin resistance in
acute and episodic care rather than chronic Asian Indian men. J Clin Endocrinol Metab.
care. The effectiveness of team-based chronic 1999;84:13744.
9. Zargar AH, Wani AI, Masoodi SR, et al. Mortality in
care management is well established but not diabetes mellitusdata from a developing region of
widely implemented. Thus, the challenge lies the world. Diabetes Res Clin Pract. 1999;43:6774.
in designing alternative care models that iden- 10. Mohan V, Shanthirani CS, Deepa M, et al. Mortality
tify people with undetected diabetes, define rates due to diabetes in a selected urban south Indian
population the Chennai Urban Population Study
individual needs, provide interdisciplinary care (CUPS-16). J Assoc Phys India. 2006;54:1137.
and measure effectiveness to make diabetes 11. Rema M, Deepa R, Mohan V. Prevalence of retinop-
prevention and control programmes accessible, athy at diagnosis among type 2 diabetic patients
affordable and sustainable. It is time to evalu- attending a diabetic centre in South India. Br
J Ophthalmol. 2000;84:105860.
ate existing policies to address diabetes and 12. Ramachandran A, Snehalatha C, Satyavani K, et al.
devise a strategy and accountability framework Prevalence of vascular complications and their risk
for short-, medium- and long-term solutions to factors in type 2 diabetes. J Assoc Physicians India.
address the growing unmet needs in diabetes 1999;47:11526.
13. Tripathy BB, Panda NC, Tej SC, et al. Survey for
prevention and control. Immediate action is detection of glycosuria, hyperglycaemia and diabe-
needed to avert this escalating health disaster. tes mellitus in urban and rural areas of Cuttack dis-
trict. J Assoc Physicians India. 1971;19:681.
14. Ahuja MMS, Sivaji L, Garg VK, et al. Prevalence of
diabetes in northern India (Delhi area). Horn Metab
Res. 1974;4:321.
References 15. Gupta OP, Joshi MH, Dave SK. Prevalence of diabe-
tes in India. Adv Metab Disord. 1978;9:14765.
1. International Diabetes Federation. Diabetes Atlas. 7th 16. Murthy PD, Pullaiah B, Rao KV. Survey for detection
ed. 2015 Update. of hyperglycaemia and diabetes mellitus in Tenali. In:
resources/2015-atlas.html. Last accessed on 21 Apr Bajaj JS, editor. Diabetes mellitus in developing
2016. countries. New Delhi: Interprint; 1984. p. 55.
2. Misra A, Sharma R, Gulati S. Consensus dietary guide- 17. Patel JC. Prevalence of hypertension and diabetes
lines for healthy living and prevention of obesity, the mellitus in a rural village. J Diabet Assoc India.
metabolic syndrome, diabetes, and related disorders in 1986;26:68.
Asian Indians. Dia Tec Ther. 2011;13(6):68394. 18. Ramachandran A, Jali MV, Mohan V, et al. High
3. NCD Risk Factor Collaboration (NCD-RisC). prevalence of diabetes in an urban population in
Worldwide trends in diabetes since 1980: a pooled South India. Br Med J. 1988;297:58790.
analysis of 751 population-based studies with 44 mil- 19. Deurenberg P, Deurenberg-Yap M, Guricci S. Asians
lion participants. Lancet. 2016;387:151330. are different from Caucasians and from each other in
their body mass index/body fat per cent relationship.
Obes Rev. 2002;3:1416.
20. Rao PV, Ushabala P, Seshiah. The Eluru survey:
Further Reading prevalence of known diabetes in a rural Indian popu-
lation. Diabetes Res Clin Pract. 1989;7:2931.
4. Sicree R, Shaw J, Zimmet P. Diabetes and impaired 21. McKeigue PM, Shah B, Marmot MG. Relation of
glucose tolerance in India. In: Gan D, editor. Diabetes central obesity and insulin resistance with high
atlas. Belgium: International Diabetes Federation; diabetes prevalence and cardiovascular risk in South
2006. p. 15103. Asians. Lancet. 1991;337:3826.
5. Mohan V, Sandeep S, Deepa R, et al. Epidemiology of 22. Wander GS, Khurana SB, Gulati R, et al.
type 2 diabetes: Indian scenario. Indian J Med Res. Epidemiology of coronary heart disease and risk fac-
2007;125:21730. tors in a rural Punjab population: prevalence and
6 Diabetes in India and Southeast Asia 97

correlation with various risk factors. Ind Heart 36. Reddy KS, Prabhakaran D, Chaturvedi V, on behalf
J. 1994;46:31923. of the Sentinel Surveillance System for Indian
23. Ramachandran A, Snehalatha C, Dharmaraj D, et al. Industrial Populations Study Group, et al. Methods
Prevalence of glucose intolerance in Asian Indians. for establishing a surveillance system for cardiovas-
Urbanrural difference and significance of upper cular diseases in Indian industrial populations. Bull
body adiposity. Diabetes Care. 1992;15:1348. WHO. 2006;84:4619.
24. Ramankutty V, Joseph A, Soman CR. High preva- 37. Deo SS, Zantye A, Mokal R, et al. To identify the
lence of type 2 diabetes in an urban settlement in risk factors for high prevalence of diabetes and
Kerala, India. Ethn Health Med. 1999;4:2319. impaired glucose tolerance in Indian rural
25. Zargar AH, Khan AK, Masoodi SR, et al. Prevalence population. Int J Diabetes Dev Countries. 2006;26:
of type 2 diabetes mellitus and impaired glucose tol- 1923.
erance in the Kashmir Valley of the Indian subconti- 38. Menon VU, Kumar KV, Gilchrist A, et al. Prevalence
nent. Diabetes Res Clin Pract. 2000;47:135. of known and undetected diabetes and associated
26. Ramachandran A, Snehalatha C, Kapur A, Diabetes risk factors in central Kerala ADEPS. Diabetes Res
Epidemiology Study Group in India (DESI), et al. Clin Pract. 2006;74:289.
High prevalence of diabetes and impaired glucose 39. Chow CK, Raju PK, Raju R, et al. The prevalence
tolerance in India: National Urban Diabetes Survey. and management of diabetes in rural India. Diabetes
Diabetologia. 2001;44:1094. Care. 2006;29:17178.
27. Misra A, Pandey RM, Devi JR, et al. High preva- 40. Raghupathy P, Antonisamy B, Fall CH, et al. High
lence of diabetes, obesity and dyslipidaemia in urban prevalence of glucose intolerance even among
slum population in northern India. Int J Obes. young adults in south India. Diabetes Res Clin Pract.
2001;25:17229. 2007;77:26979.
28. Mohan V, Shanthirani CS, Deepa R. Glucose intoler- 41. Mohan V, Mathur P, Deepa R, et al. Urban rural dif-
ance (Diabetes and IGT) in a selected South Indian ferences in prevalence of self reported diabetes in
population with special reference to family history, India-The WHO-ICMR Indian NCD risk factor sur-
obesity and lifestyle factors- the Chennai Urban veillance. Diabetes Res Clin Pract.
Population Study (CUPS 14). J Assoc Physicians 2008;80:15968.
India. 2003;51:771. 42. Ramachandran A, Mary S, Yamuna A, et al. High
29. Sadikot SM, Nigam A, Das S, et al. The burden of prevalence of diabetes and cardiovascular risk fac-
diabetes and impaired glucose tolerance in India tors associated with urbanization in India. Diabetes
using the WHO 1999 criteria: prevalence of diabetes Care. 2008;31:8938.
in India study (PODIS). Diabetes Res Clin Pract. 43. Rema M, Ponnaiya M, Mohan V. Prevalence of reti-
2004;66:3017. nopathy in non insulin dependent diabetes mellitus
30. Gupta A, Gupta R, Sarna M, et al. Prevalence of dia- in southern India. Diabetes Res Clin Pract. 1996;24:
betes, impaired fasting glucose and insulin resis- 2936.
tance syndrome in an urban Indian population. 44. Dandona L, Dandona R, Naduvilath TJ, et al.
Diabetes Res Clin Pract. 2003;61:69. Population based assessment of diabetic retinopathy
31. Agrawal RP, Singh G, Nayak KC, et al. Prevalence in an urban population in southern India. Br
of diabetes in camel milk consuming RAICA Rural J Ophthalmol. 1999;83:93740.
Community of North West Rajasthan. Int J Diabetes 45. Narendran V, John RK, Raghuram A, et al. Diabetic
Dev Countries. 2004;24:10914. retinopathy among self reported diabetics in southern
32. Ramachandran A, Snehalatha C, Baskar AD, et al. India: a population based assessment. Br
Temporal changes in prevalence of diabetes and J Ophthalmol. 2002;86:10148.
impaired glucose tolerance associated with lifestyle 46. Rema M, Premkumar S, Anitha B, et al. Prevalence
transition occurring in the rural population in India. of diabetic retinopathy in Urban India: the Chennai
Diabetologia. 2004;47:8605. Urban Rural Epidemiology Study (CURES) Eye
33. Mohan V, Deepa M, Deepa R, et al. Secular trends in Study-1. Invest Ophthalmol Vis Sci.
the prevalence of diabetes and glucose tolerance in 2005;46:232833.
urban South India-the Chennai Urban Rural 47. John L, Sundar Rao PSS, Kanagasabhapathy
Epidemiology Study (CURES- 17). Diabetologia. AS. Prevalence of diabetic nephropathy in non insu-
2006;49:1175. lin dependant diabetes mellitus. Indian J Med Res.
34. Basavanagowdappa H, Prabhakar AK, Prasannaraj 1991;94:249.
P, Gurudev KC, Virupaksha S. Study of prevalence 48. Gupta DK, Verma LK, Khosla PK, et al. The preva-
of diabetes mellitus and impaired fasting glucose in lence of microalbuminuria in diabetes: a study from
a rural population. Int J Diabetes Dev Countries. north India. Diabetes Res Clin Pract. 1991;12:1258.
2005;25:98101. 49. Yajnik CS, Naik SS, Raut KN, et al. Urinary albumin
35. Prabhakaran D, Shah P, Chaturvedi V, et al. excretion rate (AER) in newly-diagnosed type 2
Cardiovascular risk factor prevalence among men in Indian diabetic patients is associated with central
a large industry of northern India. Natl Med J India. obesity and hyperglycaemia. Diabetes Res Clin
2005;18:5965. Pract. 1992;17:5560.
98 S.R. Joshi and S.R. Aravind

50. Vijay V, Snehalatha C, Ramachandran A, et al. practices (DIPPAP-1 study). Int J Diabetes Dev
Prevalence of proteinuria in non-insulin dependent Countries. 1997;17:212.
diabetes. J Assoc Physicians India. 1994;42: 64. Raheja BS, Kapur A, Bhoraskar A, et al. DiabCare
7924. AsiaIndia Study: diabetes care in Indiacurrent
51. Mohan V, Meera R, Premalatha G, et al. Frequency status. J Assoc Physicians India. 2001;49:71722.
of proteinuria in type 2 diabetes mellitus seen at a 65. American Diabetes Association (ADA)
diabetes centre in Southern India. Postgrad Med Recommendations Regarding Glycated Hemoglobin
J. 2000;76:56973. Standardization. American diabetes association
52. Varghese A, Deepa R, Rema M, et al. Prevalence of position statement. Tests of glycemia in diabetes.
microalbuminuria in type 2 diabetes mellitus at a Diabetes Care. 2004;27:S913.
diabetes centre in Southern India. Postgrad Med 66. Dilley J, Ganesan A, Deepa R, et al. Association of
J. 2001;77:399402. A1C with cardiovascular disease and metabolic syn-
53. Ranjit Unnikrishnan I, Rema M, Pradeepa R, et al. drome in Asian Indians with normal glucose toler-
Prevalence and risk factors of diabetic nephropathy ance. Diabetes Care. 2007;30:152732.
in an Urban South Indian population. The Chennai 67. Phillips LS, Branch WT, Cook CB, et al. Clin Inertia
Urban Rural Epidemiology Study (CURES-45). Ann Intern Med. 2001;135:82534.
Diabetes Care. 2007;30(8):201924. 68. Couzin J. Clinical research: deaths in diabetes trial
54. Mohan V, Premalatha G, Sastry NG. Ischaemic heart challenge a long-held theory. Science. 2008;
disease in south Indian NIDDM patients a clinic 884885:15.
based study on 6597 NIDDM patients. Int J Diabetes 69. Kapur A. Economic analysis of diabetes care. Indian
Dev Countries. 1995;15:647. J Med Res. 2007;125:47382.
55. Ramachandran A, Snehalatha C, Latha E, et al. 70. Nagpal J, Bhartia A. Quality of diabetes care in the
Clustering of cardiovascular risk factors in urban middle and high-income group populace: the Delhi
Asian Indians. Diabetes Care. 1998;21:96771. Diabetes Community (DEDICOM) survey. Diabetes
56. Mohan V, Deepa R, Shanthirani CS, et al. Prevalence Care. 2006;29:23418.
of coronary artery disease and its relationship to lip- 71. Kapur A, Shishoo S, Ahuja MMS, et al. Diabetes
ids in a selected population in South India. J Am Coll care in India: physicians perceptions, attitudes and
Cardiol. 2001;38:6827. practices. Int J Diabetes Dev Countries. 1998;18:
57. Premalatha G, Shanthirani CS, Deepa R, et al. 12430. 72.
Prevalence and risk factors of peripheral vascular 72. Chuang LM, Tsai ST, Huang BY, Tai TY. The status
disease in a selected south Indian population the of diabetes control in Asiaa cross-sectional survey
Chennai Urban Population Study (CUPS). Diabetes of 24 317 patients with diabetes mellitus in 1998.
Care. 2000;23:1295300. Diabet Med. 2002;19:97885.
58. Ashok S, Ramu M, Deepa R, et al. Prevalence of 73. Mohan V, Goldhaber-Fiebert JD, Radha V, et al.
neuropathy in type 2 diabetic patients attending a Screening with OGTT alone or in combination with
diabetes centre in south India. J Assoc Physicians the Indian diabetes risk score or genotyping of
India. 2002;50:54650. TCF7L2 to detect undiagnosed type 2 diabetes in
59. Pradeepa R, Rema M, Vignesh J, et al. Prevalence Asian Indians. Indian J Med Res. 2011;133:2949.
and risk factors for diabetic neuropathy in an urban 74. Engelgau MM, Narayan KMV, Herman
south Indian population: the Chennai Urban Rural WH. Screening for type 2 diabetes. Diabetes Care.
Epidemiology Study (CURES-55). Diabet Med. 2000;23(10):156380.
2008;25:40712. 75. Joshi SR, Das AK, Vijay VJ, Mohan V. Challenges
60. Mohan V, Ravikumar R, Shanthirani S, et al. Intimal in diabetes care in India: sheer numbers, lack of
medial thickness of the carotid artery in south Indian awareness and inadequate control. J Assoc
diabetic and non diabetic subjects: the Chennai Physicians India. 2008;56:44350.
Urban Population Study (CUPS). Diabetologia. 76. Misra A, Nigam P, Hills AP, et al. Consensus physi-
2000;43:4949. cal activity guidelines for Asian Indians. Diabetes
61. Deepa M, Deepa R, Shanthirani CS, et al. Awareness Technol Ther. 2012;14(1):8398.
and knowledge of diabetes in Chennai the Chennai 77. Cuff DJ, Meneilly GS, Martin A, et al. Effective
Urban Rural Epidemiology Study [CURES-9]. exercise modality to reduce insulin resistance in
J Assoc Physicians India. 2005;53:2837. women with type 2 diabetes. Diabetes Care. 2003;
62. Murugesan N, Snehalatha C, Shobhana R, et al. 26(11):297782.
Awareness, about diabetes and its complications in 78. Ishii T, Yamakita T, Sato T, et al. Resistance training
the general and diabetic population in a city in improves insulin sensitivity in NIDDM subjects
southern India. Diabetes Res Clin Pract. 2007;77: without altering maximal oxygen uptake. Diabetes
4337. Care. 1998;21(8):13535.
63. Kapur A, Shishoo S, Ahuja MMS, et al. Diabetes 79. Silverstein J, Klingensmith G, Copeland K, et al.
care in India: patients perceptions attitudes and Care of children and adolescents with type 1 diabetes:
6 Diabetes in India and Southeast Asia 99

a statement of the American Diabetes Association. onset type 2 diabetes in India: CINDI 2. Indian
Diabetes Care. 2005;28(1):186212. J Endocrinol Metab. 2016;20(1):1148.
80. Joshi SR, Bhansali A, Bajaj S, et al. Results from a 94. Sosale A, Prasanna Kumar KM, Sadikot SM.
dietary survey in an Indian T2DM population: a (CINDI). Chronic complications in newly diagnosed
STARCH study. BMJ Open. 2014;4(10):e005138. patients with Type 2 diabetes mellitus in India.
81. Sadikot SM, Nigam A, Das S, et al. Diabetes India. Indian J Endocrinol Metab. 2014;18(3):35560.
The burden of diabetes and impaired fasting glucose 95. Krug EG. Trends in diabetes: sounding the alarm.
in India using the ADA1997 criteria: prevalence of Lancet. 2016;387:14856.
diabetes in India study (PODIS). Diabetes Res Clin 96. Madhu SV, et al. Guideline development group. Int
Pract. 2004;66:29330. J Diabetes Dev Countries. 2015;35 Suppl 1:S171.
82. Research India Diabetes(ICMR-INDAB) study 97. Nanditha A, et al. Diabetes in Asia and the Pacific:
(phase I): Indian Council of Medical Research India implications for the global epidemic. Diabetes Care.
Diabetes 4. Indian J Endocrinol Metab. 2016;39(3):47285.
2014;18:37985. 98. Shen J, Prabhakaran D, Tandon N, et al. A multieth-
83. The Diabetes Prevention Program Research Group. nic study of pre-diabetes and diabetes in LMIC. Glob
The diabetes prevention program. Diabetes Care. Heart. 2016;11(1):6170.
2002;25:216571. 99. Ramachandran A, et al. Combining fasting plasma
84. Lindstorm J, Louheranta A, Mannelin M, et al. The glucose with gamma-glutamyl transferase improves
Finnish diabetes prevention study. Diabetes Care. the sensitivity to predict incident diabetes in Asian
2003;26:32306. Indian men with impaired glucose tolerance. J Assoc
85. Misra A, Alappan NK, Vikram N. Effect of super- Physicians India. 2014;62(11):1822.
vised progressive resistance-exercise training proto- 100. Vinitha R, Johnston DG, Ramachandran A, et al.
col on insulin sensitivity, glycemia, lipids, and body Adiponectin, leptin, interleukin-6 and HbA1c in the
composition in Asian Indians with type 2 diabetes. prediction of incident type 2 diabetes: a nested case-
Diabetes Care. 2008;31(7):12827. control study in Asian Indian men with impaired glu-
86. Thangasami SR, Chandani AL, Thangasami cose tolerance. Diabetes Res Clin Pract. 2015;109(2):
S. Emphasis of yoga in the management of diabetes. 3406.
J Diabetes Metab. 2015;6:10. 101. Ram J, Snehalatha C, Ramachandran A, et al. Retinol
87. Youngwanichsetha S, Phumdoung S, binding protein-4 predicts incident diabetes in Asian
Inqkathawornwong T. The effects of mindfulness Indian men with prediabetes. Biofactors.
eating and yoga exercise on blood sugar levels of 2015;41(3):1605.
pregnant women with gestational diabetes mellitus. 102. Ramachandran A, et al. Improvement in diet habits,
Appl Nurs Res. 2014;27(4):22730. independent of physical activity helps to reduce
88. The DCCT Research Group. The effect of intensive incident diabetes among prediabetic Asian Indian
treatment of diabetes on the development and pro- men. Diabetes Res Clin Pract. 2014;106(3):4915.
gression of long-term complications in insulin 103. Ramachandran A, et al. Hypertriglyceridaemic waist
dependent diabetes mellitus. N Engl J Med. 1993; phenotype as a simple predictive marker of incident
329:97786. diabetes in Asian-Indian men with prediabetes.
89. Okhubo Y, Hideki K, Araki E, et al. Intensive insulin Diabet Med. 2014;31(12):15429.
therapy prevents the progression of diabetic micro- 104. Ramachandran A, et al. Effectiveness of mobile
vascular complications in Japanese patients with phone messaging in prevention of type 2 diabetes by
non-insulin dependent diabetes mellitus. A random- lifestyle modification in men in India: a prospective,
ized prospective six year study. Diabetes Res Clin parallel-group, randomised controlled trial. Lancet
Pract. 1995;28:10317. Diabetes Endocrinol. 2013;1(3):1918.
90. UKPDS Study Group. Intensive blood glucose control 105. Ramachandran A, et al. Predictive value of HbA1c
with SU and insulin compared with conventional for incident diabetes among subjects with impaired
treatment and risk of complications in patients with glucose tolerance analysis of the Indian Diabetes
type 2 diabetes. (UKPDS 33). Lancet. Prevention Programmes. Diabet Med. 2012;29(1):
1998;352:83753. 948.
91. Pung OJ, Sobieraj DM, Engel SS. Early combination 106. Anjana RM, ICMRINDIAB Collaborative Study
therapy for the treatment of type 2 diabetes mellitus: Group, et al. Prevalence of diabetes and prediabetes
systematic review and meta-analysis. Diabetes Obes (impaired fasting glucose and/or impaired glucose
Metab. 2014;16(5):4107. tolerance) in urban and rural India: phase I results of
92. Screening for Diabetes. American diabetes associa- the Indian Council of Medical Research-INdia
tion. Diabetes Care. 2002;25(1):s214. DIABetes (ICMR-INDIAB) study. Diabetologia.
93. Sosale B, Sosale AR, Mohan AR. (CINDI 2). 2011;54(12):30227.
Cardiovascular risk factors, micro and macrovascu- 107. Ramachandran A, Indian Diabetes Prevention
lar complications at diagnosis in patients with young Programme (IDPP), et al. The Indian Diabetes
100 S.R. Joshi and S.R. Aravind

Prevention Programme shows that lifestyle Indian Diabetes Prevention Programme (IDPP).
modification and metformin prevent type 2 diabe- Diabetes Care. 2007;30(10):254852.
tes in Asian Indian subjects with impaired glucose 110. Snehalatha C, et al. Beneficial effects of strategies
tolerance (IDPP-1). Diabetologia. for primary prevention of diabetes on cardiovascular
2006;49(2):28997. risk factors: results of the Indian Diabetes Prevention
108. Ramachandran A, et al. Persistent impaired glucose Programme. Diabetes Vasc Dis Res. 2008;5(1):
tolerance has similar rate of risk factors as for diabe- 259.
tes results of Indian diabetes prevention programme 111. Unnikrishnan R, Anjana RM, Mohan V. Diabetes
(IDPP). Diabetes Res Clin Pract. 2006;73(1):1003. mellitus and its complications in India. Nat Rev
109. Ramachandran A, et al. Cost-effectiveness of the Endocrinol. 2016. doi:10.1038/nrendo.2016.53.
interventions in the primary prevention of diabetes [Epub ahead of print]. Online ahead of print.
among Asian Indians: within-trial results of the
Diabetes in Latin America
Omar Y. Bello-Chavolla
and Carlos A. Aguilar-Salinas

Classication and Unique Aspects areas of Argentina or Uruguay. In urban areas of

of the Pathophysiology of Type 1 Colombia and Mexico, the average proportion of
and Type 2 Diabetes in the Region Native American and European admixture are
5060 % and 3045 %, respectively. In addition,
The Latin America (LA) region is comprised of an ongoing demographic transition has modified
21 countries, including Mexico, six countries in the age structure of Latin American populations
Central America, ten countries in South America, with important increases in the adult (15
and three countries in the Caribbean islands span- 59 years) population and especially for popula-
ning a total population of almost 600 million tions of older populations (aged 60 years and
people. The region is composed of low- and over) [1].
middle-income countries with populations The rapid growth of the number of inhabitants
mainly composed of young adults. Ethnic diver- in the Latin American population observed in the
sity and the preservation of the Amerindian heri- first half of the preceding century was reverted in
tage are unique features of the Latin American the past few decades (2.8 % growth per year in
populations. Meanwhile, the Amerindian compo- 1950, 1.3 % in the early 2000s). This trend
nent is the dominant genetic background in Peru resulted from birth control campaigns, migratory
and Guatemala; its contribution is minimal in movements, and economic phenomena. On the
other hand, a trend for decreasing all-cause mor-
tality started to occur within Latin America
50 years ago. Life expectancy in the region has
increased by 21.6 years on average reaching
73.4 years for the 20052010 period. The varia-
O.Y. Bello-Chavolla, MD, PhD (PECEM) tion of life expectancy is wide within the region,
Department of Endocrinology and Metabolism, ranging from 65.5 in Bolivia to >78 years in
Instituto Nacional de Ciencias Mdicas y Nutricin Chile and Costa Rica. As a result, the percentage
Salvador Zubirn, Mexico City, Mexico of elder population has increased in the last
Facultad de Medicina, Universidad Nacional 50 years. It is expected that, by the year 2050, the
Autnoma de Mxico, Mexico City, Mexico percentage of elders move from 8.8 to 23.6 %.
C.A. Aguilar-Salinas (*) The demographic modifications pose a remark-
Departamento de Endocrinologia y Metabolismo, able challenge for the local healthcare systems.
Instituto Nacional de Ciencias Mdicas y Nutricion,
Vasco de Quiroga 15, Mexico City 14000, Mexico The increase of life expectancy will lead to
e-mail: increased number of susceptible cases for having

Springer International Publishing Switzerland 2017 101

S. Dagogo-Jack (ed.), Diabetes Mellitus in Developing Countries and Underserved Communities,
DOI 10.1007/978-3-319-41559-8_7
102 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

non-transmissible chronic diseases (i.e., type 2 triglycerides, and abnormal fasting glucose.
diabetes mellitus), which already are the major When three or more of these abnormalities are
health burden within Latin American countries. present, individuals are diagnosed as having the
metabolic syndrome. This condition is associated
with a fivefold increase for having T2D. Several
Type 2 Diabetes metabolic syndrome definitions have been pro-
posed. Using the ATP-III criteria, the age-
Close to 80 % of 415 million T2DM cases live in adjusted prevalence for the region is 28.3 %.
middle- and low-income countries. A significant Population-based data are available for Brazil
proportion of them (41.1 millions) reside in (29 %), Peru (18.1 %), Colombia (34.8 %),
LA. This number is expected to increase to 69.4 in Venezuela (35.3 %), and Mexico (36.8 %). Higher
2040. The International Diabetes Federation (IDF) percentages are found if the IDF definition is
estimated that the age-adjusted prevalence for the applied (for Mexican adults, the prevalence
region was 9.2 % for adults (aged 2079 years) in changes from 36.8 to 49.8 %). The prevalence of
2014. Only North America (10.5 %) and southern the metabolic syndrome traits is different in LA
Asia (10.9 %) had a greater prevalence of diabetes compared to that reported in Caucasian popula-
than LA. Two of the ten leading countries for the tions; low HDL cholesterol, abdominal obesity,
number of cases are located in the region (Brazil and hypertriglyceridemia are significantly more
(14.3 million) and Mexico (11.5 million)) common in LA subjects.
(Table 7.1). In addition, the increment in the num- The clinical expression of T2D has several
ber of cases is expected to be greater in this region peculiarities that have been consistently reported
compared to other areas. in LA populations. Especially among mestizo
According to the Global Burden of Disease individuals, the disease is expressed at an earlier
group, the mean fasting plasma glucose of the age and with a lower body mass index (BMI)
adults living in LA has increased 3.06 mg/dl compared to that reported in Caucasians. The
every decade since 1980 [2]. The end result has premature age of onset increases the social and
been a rapidly increasing prevalence of diabetes. economic burden because of the higher preva-
The epidemic growth is expected to continue lence of chronic complications and premature
since the prevalence of several preceding condi- disability during productive years. In Mexico,
tions of T2D (i.e., obesity and glucose intoler- 22.7 % of people with T2D are under the age 40.
ance) is higher in LA than in other regions. Individuals with early-onset T2D comprise a
According to the latest report by the Food and heterogeneous population [4]. Two thirds of
Agriculture Organization of the United Nations them have BMI >25 kg/m2; these cases had an
(FAO) [3] and the World Health Organization increased prevalence of several metabolic syn-
(WHO), the worldwide prevalence of obesity in drome traits (e.g., hypertension and hypoalph-
2013 was 11.7, whereas the prevalence of obesity alipoproteinemia) and can be controlled using
for the Latin American was reported to be 23.4 %, oral glucose-lowering agents. In contrast, a
distributed in 30.4 % for Central America and large proportion of the lean cases require insulin
21.6 % for South America (Table 7.2). The coun- as part of their treatment. Compared to the over-
tries with the highest obesity prevalence in the all population with T2D, the young T2D popu-
LA region are Mexico, followed by Venezuela, lation had a higher prevalence of underdiagnosis,
Argentina, and Chile. The reported prevalence more school years but a lower socioeconomic
for obesity in Mexican women proves to be level, and higher alcohol and tobacco consump-
higher than the same figure reported for the USA tion. Few young patients undertake preventive
in 2008 (31.8 %). A high percentage of patients measures, with very few receiving statins, ace-
with excess body weight have central fat accumu- tylsalicylic acid, or follow-up with an ophthal-
lation, arterial hypertension as well as abnormal mologist. This group has specific barriers to
concentrations of total cholesterol, HDL and adhere treatment programs (e.g., depression,
7 Diabetes in Latin America 103

Table 7.1 Prevalence of diabetes in Latin America

Annual death Annual Estimated Annual
rate due to treatment cost number of increment of
Number of cases diabetes per subject undiagnosed the number of
Country (2079 years) Prevalence (2079 years) (US dollars) cases cases
Argentina 1,570,200 5.57 15,416 966.44 722,290 29,000
Bolivia 325,220 6.89 4,732 124.63 149,600
Brazil 14,300,000 10.52 129,226 1,031.44 6,164,590 377,000
Chile 1,442,610 12.78 10,459 992.13 755,600 28,000
Colombia 2,067,870 7.26 14,602 482.72 951,220 95,000
Costa Rica 259,350 8.81 1,659 937.20 119,300
Cuba 872,950 8.58 7,560 823.71 401,560 19,000
Ecuador 563,840 6.89 5,492 335.41 259,360 19,000
El Salvador 312,430 9.88 3,233 333.58 143.72
French Guiana 12,610 9.60 5,800
Guatemala 589,140 9.93 7,202 311.52 271,010 27,000
Honduras 239,590 7.16 2,338 209.40 110,210
Mexico 11,500,000 14.4a 80,000a 815.53 3,452,410 323,000
Nicaragua 309,320 11.58 3,001 172.21 142,290
Panama 184,580 8.59 1,399 732.45 84,910
Paraguay 222,220 6.81 2,174 283.14 102,220
Peru 1,108,610 6.81 8,150 307.31 509,960
Puerto Rico 391,870 12.98 108,590
Dominican 405,580 7.36 5,183 419.28 186,570
Uruguay 157,330 6.02 1,122 922.68 72,370
Venezuela 1,764,900 10.39 13,380 914.01 811,850 61,000
Modified from: Whiting et al. [23]. International Diabetes Federation Atlas 2012
Villalpando et al. [93]

work-related stress, and alcoholism) that should the presence of arterial hypertension. Smoking
be intentionally sought. persists as a common risk factor in the patients
On the other hand, the mean BMI of recently with diabetes. Dyslipidemia is one of the most
diagnosed populations is lower in LA patients common comorbidities in T2D, with higher lev-
(e.g., for Mexico, 27.9 kg/m2 in males and els of triglycerides and non-HDL cholesterol
28.9 kg/m2 in females) compared to that reported compared with values of the general population.
in the USA (usually above 30 kg/m2). Mestizo LDL cholesterol (LDL-C) levels >100 mg/dL are
populations are less tolerant to excess body observed in 74.8 % (95 % CI 72.576.9 %) of pre-
weight because fat is accumulated in ectopic viously diagnosed patients [6]. Similar percent-
organs and in the intra-abdominal cavity instead ages have been reported by the Qualidiab network
of the subcutaneous adipose tissue [5]. using data from several South American coun-
T2D coexists frequently with other comorbid- tries [7].
ities. Taking as an example the Mexican popula- A high percentage of women with T2D had at
tion, a high percentage of patients with T2D had least one pregnancy during their lifetimes
at least one cardiovascular risk factor (86.7 %) (94.7 %); this proportion was similar to the one
(e.g., hypercholesterolemia, family history of found in patients without T2D. However, the
cardiovascular mortality, arterial hypertension, number of women who had suffered at least one
and smoking). Nearly half the patients had hyper- abortion was significantly higher in the group
tension, but almost half of them are unaware of with diabetes (OR 1.62, 95 % CI 1.531.83) with
104 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

Table 7.2 Average body mass index (BMI) standardized for age by age and country for the years 1980 and 2008
Prevalence of obesity
Men (kg/m2) Women (kg/m2) in adults (%)
1980 2008 1980 2008 2008
Argentina 25.4 27.5 23.8 27.5 29.4
Bolivia 23.0 24.4 23.6 26.9 18.9
Brazil 22.6 25.8 24.1 26.0 19.5
Chile 24.5 27.0 24.0 27.9 29.1
Colombia 22.1 24.9 23.4 26.2 18.1
Costa Rica 23.7 26.5 23.4 27.0 24.6
Cuba 22.7 25.1 23.8 26.6 20.5
Ecuador 23.8 25.6 24.7 27.1 22.0
El Salvador 23.8 26.4 23.7 27.8 26.9
Guatemala 23.2 25.3 22.9 26.8 20.7
Honduras 23.1 25.1 23.0 26.7 19.8
Mexico 24.5 27.4 24.6 28.7 32.8
Nicaragua 24.0 25.8 24.4 27.6 24.2
Panama 23.5 26.3 23.9 27.7 25.8
Paraguay 23.4 25.5 23.5 25.9 19.2
Peru 23.0 24.8 25.1 26.0 16.5
Dominican Republic 22.9 25.2 22.6 26.9 21.9
Uruguay 24.2 26.4 23.3 26.6 23.6
Venezuela 24.6 27.4 24.7 28.1 30.8
Adapted from: Finucane MM, Stevens GA, Cowan MJ, et al., and the Food and Agriculture Organization of the United
Nations. The State of Food and Agriculture 2013

a similar trend found in the risk of stillbirth (OR diagnosed after the age 70; these patients have a
1.99, 95 % CI 1.752.3); these differences were low prevalence of microvascular complications
held significant when adjusted by age. A high and their glucose levels can be kept stable with
percentage of women with T2D during their one or two oral hypoglycemic agents. Both
reproductive years did not use contraceptive groups are represented in similar proportions.
methods (42.5 %); this rate was not significantly Cardiovascular risk factors were common in this
different in women without T2D (38.8 %). A age group; 60 % had hypertension, 88.7 % was
quarter of patients with T2D had a current or pre- taking one or more antihypertensive agents, and
vious history of a urinary tract infection during microalbuminuria was detected in 48.4 % of
the previous year. Additionally, the risk of suffer- cases. These patients have a higher prevalence of
ing an accident is higher for T2D patients (OR geriatric syndromes (falls, motor limitations,
1.42, 95 % CI 1.251.60) with half of the acci- cognitive dysfunction, and polypharmacy) [9].
dents occurring within the patients house. The T2DM patients in Latin America are exposed
risk of accidental falls in patients with T2D is not to endemic and highly prevalent infectious dis-
held significant when stratified by age group [8]. eases (such as tuberculosis, influenza, HIV, and
The elder patient with T2D is a growing group hepatitis C), which represent a serious comorbid-
in the region. Two major profiles are found ity within this group. T2D increases the risk for
among them. The first is composed of T2D having tuberculosis, and it is associated with a
patients with a long exposure to the disease and decreased rate of eradication. The epidemiologic
chronic complications, third-party dependence, transition from infectious to noncommunicable
and requiring a more complex management. The chronic diseases has created a unique environ-
second group is composed of T2D patients ment in Latin American countries where the
7 Diabetes in Latin America 105

interaction of both disease trends increases mor- The information about the epidemiology of
bidity and mortality associated with noncommu- T1DM in this region is scant [12].
nicable diseases [10]. In most high-income countries, the majority
of diabetes in children and adolescents is caused
by T1DM. This statement does not apply for the
Gestational Diabetes LA region. The growth in the number of children
with diabetes is mainly caused by T2D. In
Gestational diabetes is a common obstetric com- Mexico, half of the children with diabetes are
plication in the LA region. Prevalence depends obese, do not have positive titers of GAD anti-
on the sampling approach and the diagnostic cri- bodies, and are controlled with oral glucose-
teria applied. It varies between 4.3 and 30.1 %; lowering agents.
the majority of reports inform percentages close The WHO coordinated a Multinational Project
to 10 %. The use of the recently introduced for Childhood Diabetes (DIAMOND) from 1990
IADPSG criteria results in remarkably higher to 1999 in order to monitor the patterns of world-
percentages [11]. According to the 2015 IDF wide incidence of T1DM in children up to the
Atlas, the age-adjusted prevalence of hyperglyce- year 2000 [13]. The Diabetes and Genetic
mia during pregnancy is 11.5 % (crude 13.2 %) in Epidemiology Unit of the National Public Health
the LA region; it means that close to 0.9 million Institute in Helsinki, Finland served as the coor-
live births are affected by this condition. dinator unit, and 100 centers from 50 countries
Macrosomia is frequently associated with ges- around the world headed by a principal investiga-
tational diabetes. In a multinational survey, the tor were recruited. To be eligible, the center must
prevalence of macrosomia was between 2.8 and have an accurate, well-defined, population-based
9.3 % in the LA region. This percentage was registry. The inclusion criteria considered chil-
greater than that reported in Asia and Africa. dren 014 years with residency in the study area.
Macrosomia is a strong risk factor for having Fifty countries participated in this program.
cesarean section and various adverse perinatal Seventy-five million children fitted the inclusion
outcomes (e.g., fetal distress and cephalopelvic criteria and 19,164 were diagnosed with T1DM
disproportion). from 1990 to 1994, according to WHO classifica-
tion and diagnostic criteria. Ten Latin American
countries (Argentina, Brazil, Chile, Colombia,
Type 1 Diabetes Mellitus (T1D) Paraguay, Peru, Uruguay, Venezuela, Cuba, and
Mexico) participated with 13 centers. The overall
The IDF reported that 542,000 children age-adjusted incidence of T1D varied from
(<15 years) worldwide are affected by this dis- 0.1/100,000 per year in Zunyi, China, and
ease; 86,000 new cases are diagnosed every year. Venezuela to 36.8/100,000 in Sardinia and
Prevalence of T1DM has been increasing in 36.5/100,000 in Finland (350-fold variation
recent years worldwide. An annual increase in among 100 populations around the world). The
incidence of 3 % has been recorded. incidence among populations in South America
The estimated number of cases in LA is ranged from intermediate to very low. The high-
45,100; 7,300 new cases are diagnosed every est incidence rates were among European and
year. Two of the top ten countries with the highest North American populations [14, 15]. The inci-
number of cases are located in the region (Brazil dence rates for T1DM in Latin American coun-
(30,900) and Mexico (13,500)). However, none tries are lower than those described in Spain
of the countries with the highest number of new (12.4/100,000 (11.713.1)) or Portugal
cases per year are located in LA. Readers should (14.6/100,000 (10.619.6)), suggesting that the
be aware that it is likely that these numbers are Amerindian genetic background might be protec-
underestimated due to the lack of national regis- tive against T1DM development. Native unmixed
tries in the majority of the countries of the region. minority groups remain in rural areas of Mexico,
106 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

Table 7.3 Incidence of type 1 diabetes in Latin America per 100,000 persons aged 14 years or younger
Incidence (%) (95 % Annual change of Estimate of
Country Study period CI) incidence ascertainment (%)
Puerto Rico 19901999 16.8 (16.017.6) 1.0 (2.7; 0.7) 9097
Uruguay (Montevideo) 1992 8.3 (5.411.7) 97
Brazil (Sao Paulo) 19901992 8.0 (5.5311.14) 16 (48.6; 37.2) 7095
Argentina (Avellaneda) 19901996 6.3 (5.711.1) 0.4 (8.8; 10.5) 88100
Argentina (Tierra del 19931996 10.3 (5.518.5) 100
Colombia (Bogota) 1990 3.8 (2.94.9) 97
Colombia (Cali) 19951999 0.5 (0.30.7)
Chile (Santiago) 19901999 3.7 (3.44.0) 7.5 (4.3; 10.9) 100
Cuba 19901999 2.3 (2.22.5) 10.8 (13.4; 25100
Mexico (Veracruz) 19901993 1.5 (0.72.9) 100
Paraguay 19901999 0.9 (0.81.0) 0.5 (5.7; 4.9)
Peru (Lima) 19901994 0.5 (0.40.64) 12.1 (7.5; 35.8)
Dominican Republic 19951999 0.5 (0.40.7) 12.6 (11.4; 43.0)
Venezuela (Caracas) 19901994 0.1 (0.10.2) 6.8 (24.6; 15.3)
Hispanics in the USA 19901999 11.4 (10.112.9) 51100
Adapted from The DIAMOND Project Group and Gmez-Daz et al.
Missing data

Bolivia, Peru, and Guatemala; very low incidence 3,848 T2DM patients (cases) and 4,366 con-
of T1DM has been reported in these groups. trols and have Native American and European
Furthermore, T1DM prevalence seems to be ancestry [16]. The higher susceptibility of the
related to the proportion of Caucasoid popula- admixed Latin American population can be
tions in a certain country (Table 7.3). explained both by the high prevalence of
Caucasian- and ethnic-specific risk alleles.
Fifty-six of the 68 known genetic associations
Genomic Landscape of Diabetes identified in Caucasians were replicated. The
in the Region TCF7L2 and KCNQ1 risk alleles are highly
prevalent in mestizos. In addition, two ethnic-
Type 2 Diabetes specific associations were found. The strongest
novel association was found at chromosome
Amerindian populations have an increased risk 17p13.1, specifically an SLC16A11 haplotype
for having T2D. Populations with Amerindian (OR 1.29 (95 % CI 1.201.38)). Individuals
ancestry comprise 27.663 % of the genetic with the risk haplotype develop T2DM
background of the LA countries. The Slim 2.1 years earlier (p = 3.1 104) and had 0.9 kg/
Initiative in Genomic Medicine for the m2 lower BMI than noncarriers. This frequency
Americas (SIGMA) Type 2 Diabetes of this variant is very small in Europeans and
Consortium was set out to characterize the Africans, moderate in Asians (~10 %), and
genetic factors that explain the higher T2DM high in the Americas native population (50 %).
prevalence registered within this population. This newly discovered risk allele is thought to
The SIGMA consortium genotyped 8,214 be derived from Neanderthal introgression.
Mexicans and Latin Americans, which included Approximately 20 % of the difference in
7 Diabetes in Latin America 107

prevalence of T2DM among people with opportunity to identify the existence of protective
Amerindian and European ancestry can be genotypes or the absence of susceptibility vari-
explained by the presence of the risk allele. ants in the Amerindians. T1DM is a multifacto-
The second risk allele associated with T2DM rial and polygenic disease that exerts a high
in the Latin American population is the p. genetic susceptibility trait with a concordance
E508K variant of the hepatocyte nuclear factor rate in twins of about 3050 %. Genetic variants
1- (HNF-1). It is present in 2.1 % of T2DM of the human leukocyte antigen (HLA) on chro-
patients [17]. The p.E508K variant partially mosome 6p21.3, particularly combinations of
reduces transactivation activity. Although the DR3/DR4, produce the highest risk, which
affected gene is the cause of MODY (maturity- explains nearly 50 % of the genetic contribution.
onset diabetes of the young)-3, the clinical pro- Other non-HLA T1DM susceptibility genes have
file of the patient with the p.E508K variant is been identified also.
undistinguishable from T2D. Cruz-Tapias et al. published a meta-anal-
Several polymorphisms have been studied ysis designed to estimate the risk associ-
among other populations; in Brazilians the ated with variations in HLA class II in some
TCF7L2 rs7903166 (C/T) was associated with autoimmune diseases, including T1DM in
T2DM risk among the southern-Brazilian popu- Latin American countries. Major risk alleles
lation. The frequency of the minor allele was related to T1DM incidence showed to be
38 % in the type 2 diabetes group and 31 % in DQA1*301/*501, DQB1*201/*302/*301, and
nondiabetic subjects, and this allele was signifi- DQB1*401/*402/*405, while the protective
cantly associated with type 2 diabetes risk ones were DQB1*501, DQB1*602/*603, and
(OR = 1.42, 95 % CI 1.151.76 for the dominant DQB1*11/*13/*14/*15. Meanwhile, Gorodezky
model of inheritance). et al. identified HLA haplotypes with a strong
association with T1DM in Mexicans which
included DRB1*0301-DQA1*0501-DQB1*0201
Gestational Diabetes (OR = 21.4), DRB1*0405-DQA1*0301-DQB1*
0302 (OR = 44.5), and the same DQA1/DQB1
Few genetic studies have been focused in women with the HLA haplotype DRB1*0404/*0401
with gestational diabetes in LA. Recently, conferring lower risk, increasing the risk of
Chagoya and coworkers found an association an earlier age at onset (OR = 61.3). Finally, a
between gestational diabetes and two of the most meta-analysis limited to LA patients with T1D
frequently replicated T2D loci: a TCF7L2 haplo- (21 studies, 1,138 cases, and 1,920 controls)
type (rs7901695, rs4506565, rs7903146, found that DRB1*0301 (OR: 9.65; 95 % CI:
rs12243326; P = 2.16 10-06; OR = 2.95) and a 5.6916.36; p < 0.0001), DRB1*1201 (OR: 4.84;
KCNQ1 haplotype (rs2237892, rs163184, 95 % CI: 1.9711.91; p = 0.001), DQB1*0302
rs2237897; P = 1.98 10-05; OR = 0.55). This (OR: 4.58; 95 % CI: 3.366.26; p < 0.0001),
finding is in accordance with the strong relation DQA1*0301(OR: 3.02; 95 % CI: 1.376.65;
that exists between T2D and gestational p = 0.0059) and DQB1*0602 (OR: 0.19; 95 %
diabetes. CI: 0.110.33; p < 0.0001), DRB1*14 (OR: 0.18;
95 % CI: 0.060.55; p = 0.0024), and DQB1*0501
(OR: 0.47; 95 % CI: 0.260.83; p = 0.0097) were
Type 1 Diabetes the most significant alleles associated with
T1D. Despite of the above, Latin American popu-
As recognized by the Diabetes Epidemiology lations and the Amerindian individuals have been
Research International Group, the ethnic diver- underrepresented in the T1DM genetic studies.
sity of the Latin American populations yields an Future GWAS study may consider the inclusion
108 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

of ethnic diverse populations composed by both aged 2069 years old. The corresponding rate in
high (i.e., Caucasian) and low (i.e., Amerindians) Peru is 22.4 % [29]. This information is in clear
incidence groups [1820]. contrast with the remarkably lower prevalence of
impaired glucose tolerance (7.9 %, 42.2 million)
reported for the region by the IDF.
Diagnosis of Diabetes A high percentage of T2DM patients remain
and Prediabetes in the Region undiagnosed. For example, in Mexico, half of the
patients with diabetes were unaware of their con-
T2D dition. The undiagnosed proportion of cases in
the region is not different from the rest of the
Prevalence data are available for the majority of developing countries but remarkably higher com-
the countries of the region [2125]. However, the pared to some European countries (nearly 6 %).
design of the surveys is heterogeneous leading to This percentage is even higher in young adults
contrasting conclusions even in the same country (70 %) or low-income groups. Several countries
[26]. Table 7.1 shows the prevalence reported by have implemented screening programs. The strat-
IDF which is derived mainly from population- egies are diverse. For example, Brazil had a mas-
based surveys. Twelve Latin American countries sive screening program in 2001, based in capillary
have prevalence higher than the world average glucose tests (n = 22,069,905). It covered 73 % of
(8.3 %). More than 10 % of the adult population the target population [30]. In Mexico, the adult
is affected in Chile, Puerto Rico, Nicaragua, population who attended for T2D screening in
Venezuela, Mexico, and Brazil. Prevalence is the previous year increased from 10.5 % in 2000
higher in urban settings, native Amerindian to 22.7 % in 2006. However, a major limitation of
communities (>25 % in Canada and the USA), the screening programs is the lack of a systematic
low-income/low-education groups, and popula- inclusion of the detected cases into treatment
tions that have undergone migratory movements programs.
(e.g., 25.7 % in the US/Mexico border). In Incidence data are scant in the region. In the
Mexico, the mean age of onset is 48 years, being Mexico City study, Gonzalez Villalpando and
lower in women. The highest prevalence is found coworkers found an incidence rate of 1.4 cases
in the 5560-year-old group. The time since per 100,000 persons per year in low-income sub-
diagnosis is 9.3 years in males and 8.4 years in jects [31]. Meza and coworkers using a Markov
females [27]. model estimated that incidence rates may vary
There is a lack of data about the prevalence of between five and 25 cases per 100,000 persons
prediabetes in LA [28]. This term encloses three per year among adults aged 3090 years living in
conditions (i.e., impaired fasting glucose (100 Mexico during 2010 [32]. Clearly, an unmet
125.9 mg/dl), impaired glucose tolerance (2 h need in the LA area is the existence of represen-
post-challenge plasma glucose between 140 and tative cohorts of patients with and without diabe-
199 mg/dl), and abnormal HbA1c (5.7 to 6.4 %) tes with a long-term follow-up. Incidence data
associated with an increased risk for having T2D are critical to design and validate public
in the next 10 years. The prevalence of prediabe- policies.
tes in Hispanics living in the USA was 36.8 %
(95 % CI 32.141.7 %) in 20112012 T2DM in Mexico
NHANES. Regrettably, no comparable informa- Mexico has four population-based, nationwide
tion is available in the region because only fast- surveys in which the prevalence of T2D has been
ing glycemia has been registered in the majority measured (National Chronic Disease Survey
of the LA population-based surveys. In the [ENEC] in 1993, National Health Survey [ENSA]
Mexican 2006 National Health and Nutrition in 2000, and the National Health and Nutrition
Survey (ENSANUT, in Spanish), the prevalence Surveys (ENSANUT) performed in 2006 and
of impaired fasting glucose was 19.1 % in adults 2012) [3335]. The surveys (Fig. 7.1) have
7 Diabetes in Latin America 109

Fig. 7.1 Comparison of Prevalence of T2DM in Mexico

T2DM prevalence reported
16 14.4
for the National Health and
Nutrition Polls in Mexico

Prevalence (%)
for the years 1994, 2000, 9.2 9.2
2006, and 2012, comparing 10
6.7 7.3
the cases with previous 8
medical diagnosis and the 6 4 4.6
cases with both previous 4
medical diagnosis and the 2
cases found by the polls 0
(Adapted from: Hernndez- 1994 2000 2006 2012
vila et al.) Year of ENSANUT

Previous Medical Diagnosis Previous Medical Diagnosis + Cases Found in Poll

proved an increase in prevalence from 6.7 % in and 2.86 per 100 person-years was obtained for
1993 (previously diagnosed (PD) 4.6 % and undi- men and women, respectively, despite taking into
agnosed (UD) 2.1 %) to 7.5 % in 2000 (PD 5.8 % consideration different diabetes-associated risk
and UD 1.7 %) to 14.4 % in 2006 (PMD 7.3 % factors. In addition, Mexico City was among the
and 7.1 % FP). The increases were similar for cities included in the CARMELA study, a multi-
both sexes and for rural and urban areas. The national survey designed to assess the prevalence
growing trend in T2DM prevalence is multifacto- of diabetes, dyslipidemias, and other cardiovas-
rial; aging of the population, the large proportion cular risk factors. Mexico City had the highest
of Amerindian ethnic background, and an prevalence of T2D in the region.
increase in the prevalence of obesity attributable
to changes in the lifestyle are the most obvious T2DM in Brazil
explanations. Results from ENSANUT 2012 According to data from the IDF, Brazil ranks
show that the prevalence of T2DM by PD is 9.2 % fourth among countries with the largest number
in adults over 20 years of age; this implies that of people with T2DM, comprising nearly 14.3
6.4 million Mexican adults have the diagnosis of million cases; the T2DM prevalence in Brazil has
T2DM which shows an overall doubling up from been estimated by the IDF to be 10.52 % in 2012.
the prevalence recorded in the year 2000. A There is significant variability between regions
recent IDF report estimates that 11.5 million ranging from 5.2 % (Brasilia) to 13.5 % (So
(95 % CI 6.213.7 million) Mexican adults are Carlos). The Surveillance of Risk and Protective
affected. Factors for Chronic Diseases Telephone Survey
Several complimentary sources of informa- (VIGITEL, in Portuguese) is a telephone surveil-
tion should be highlighted. The diabetes study of lance system to identify risk and protective fac-
Mexico City is a population-based cohort study tors for noncommunicable chronic diseases in
of subjects from six low-income colonies located subjects aged 18 years or older; through this sys-
within the periphery of the ABC hospital in tem, a trend of increasing prevalence rates for all
Mexico City [36]. The recruitment of adults with- NCCDs has been identified. In 2013, VIGITEL
out T2DM diagnosis started in 1990 reaching reported a prevalence of self-reported T2DM of
1,754 subjects with an average follow-up period 6.9 %. The trend of increasing prevalence has
of 11 years; after the follow-up period, an inci- also been demonstrated for different age groups
dence of 1.42 per 100 person-years for men and as shown in Fig. 7.2, where the prevalence ranges
1.21 per 100 person-years for women was from 8.5 % in the 4554-year group to 17.1 % in
obtained. The resulting incidence was inferior to the 5564-year group and 22.1 % in the groups
those obtained by a similar study undertaken in with 65 years or older. The rates of self-reported
San Antonio, TX, in which an incidence of 2.7 diabetes through VIGITEL have also shown a
110 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas


self-reported T2DM

prevalence trends from 2006

to 2013 in the Brazilian


population as reported by the

VIGITEL strategy (Adapted

from Almeida-Pititto et al.)

2006 2007 2008 2009 2010 2011 2012 2013

difference in its distribution: northern Brazil, concentrating in urban areas, along with nutri-
characterized by lower incomes, reports T2DM tional changes and population aging, has contrib-
rates of 3.65.5 %, whereas southern Brazil, with uted to an accumulation of cardiovascular risk
higher incomes, reports rates ranging from 6.7 to factors that have led to a general increase in the
8.2 %. The rates of T2DM have been reported to prevalence of T2DM [3739].
be higher specifically in Japanese Brazilians and
in Native American groups (28 % of Xavantes has T2DM in Central America
T2D). The epidemiologic transition from infectious
Obesity rates in Brazil have similarly been diseases to noncommunicable chronic diseases
escalating across all age groups; serial data from has also changed the epidemiologic landscape
the Household Budget Survey showed that BMI within Central America. The top three countries
>25 kg/m2 have increased from 16 to 50 % in men with the highest prevalence are Nicaragua,
and from 28 to 48 % in women. Likewise, the Guatemala, and El Salvador [40]. The Central
prevalence of BMI >30 kg/m2 has increased from American Diabetes Initiative (CAMDI) 2010 is
8.9 to 12.5 % in men and from 13.1 to 16.9 % in a study by the Pan American Health Organization
women. Other risk factors include physical inac- with the objective to determine the prevalence
tivity and sedentary lifestyle; in Brazil nearly of T2DM and hypertension in people 20 years
41 % of the adult population is not sufficiently or older (n = 10,822) in a sample taken from six
active to achieve health benefits. VIGITEL has Central American populations (urban areas of
reported the prevalence of T2DM risk factors as San Jos, Costa Rica; Santa Tecla, San Salvador,
follows: high frequency of soft drink consump- El Salvador; Villanueva, Guatemala City,
tion (23.3 %), alcohol abuse (16.4 %), cigarette Guatemala; Tegucigalpa, Honduras; Managua,
smoking (11.3 %), and physical inactivity Nicaragua; and the national population of
(33.8 %). The health system provides the full Belize) [41]. Prevalence of previously diag-
range of services for diabetes care and prevention nosed T2DM was similar among males and
but not universally. The federal government cov- females (4.9 vs 5.3 %); the overall prevalence of
ers 5080 % of costs, including basic medicines diagnosed hypertension was higher among
under prescription and testing strips for people women (19.0 %. 95 %-CI = 16.921.4) than
with T1DM. Similar to what has been happening among men (10.9 %, 95 %-CI = 9.612.5).
in other Latin American countries, the epidemio- Overall 5.1 % of participants reported diagnosed
logic transition motivated by populations diabetes while 3.4 % were found to have newly
7 Diabetes in Latin America 111



10.5 Overall Men Women












Fig. 7.3 Comparative prevalence of the sites studied by Survey of Diabetes, Hypertension and Chronic Disease
CAMDI, comparing the overall prevalence with preva- Risk Factors. Belize, San Jos, San Salvador, Guatemala
lence among men, women, and overall (Adapted from: City, Managua, Tegucigalpa, and Washington, D.C.:
The Central America Diabetes Initiative (CAMDI): PAHO, 2011)

diagnosed diabetes for a total prevalence of T2DM in South America

8.5 %. An additional 18.6 % was reported to After Brazil, the countries with the highest preva-
have prediabetes (IFG and/or IGT). The preva- lence are Colombia, Chile, Argentina, and
lence of previously or newly diagnosed diabetes Venezuela. The CARMELA study, conducted in
and prediabetes was comparable in men and seven cities of South America and Mexico, found
women. As shown for all participant sites in no difference in the prevalence between
Fig. 7.3, the prevalence of known and newly Barquisimeto (Venezuela), Bogot (Colombia),
diagnosed T2DM was the lowest in Tegucigalpa Quito (Ecuador), Buenos Aires (Argentina), and
(2.5 % and 2.9 %, respectively); the lowest prev- Santiago (Chile) with a significantly lower preva-
alence of prediabetes was found in Managua lence only for Lima (Peru) [42].
(12.4 %). A population-based, nationwide survey done
In the case of Guatemala for the year 2014, in Peru in 2012 found a T2D prevalence of 7 %
the IDF reports that the prevalence in adults (95 % CI 5.38.7 %). It was estimated that
aged 2079 years was 8.9 % with a total number 763,600 Peruvians may live with diabetes. The
of T2DM cases of 680,000 where nearly 107,500 prevalence of previously diagnosed cases was 4.2
cases are undiagnosed. The number of deaths and 22.4 % for impaired fasting glucose. The
attributable to T2DM in Guatemala during 2014 capital city (Lima) has a higher prevalence com-
was 7,965 implying a cost (both direct and indi- pared to the rest of Peru (8.4 vs 6 %). Peru showed
rect) of 385.4 USD. For Nicaragua in the year a twofold difference in the prevalence of T2DM
2014, the IDF reports a prevalence of T2DM in in the urban population (8.2 %) with respect to
adults aged 2079 years of 10.3 % with a total suburban areas in the mountain and in the
number of T2DM cases of 356,100, where jungle.
98,900 are undiagnosed; the number of deaths In Colombia, several cross-sectional surveys
attributable to T2DM during 2014 in Nicaragua have measured the prevalence of T2D in urban or
was 54.7 with a cost per person with diabetes of rural populations. The highest prevalence
221.3 USD. (8.93 %) was reported in Cartagena in 2006. A
112 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas


T2DM Prevalence Metabolic syndrome prevalence










Fig. 7.4 Comparative prevalence of T2DM and metabolic syndrome by Hispanic/Latino background, comparing the
overall prevalence among different ethnic backgrounds (Adapted from: Schneiderman et al. [44])

growing trend in the prevalence of the disease medical care are among the particular challenges
was observed in the time period 20092012 that should be overcome in this group [46]. The
based on the number reported by institutions of prevalence is not homogeneous between
the federal government. The magnitude of the Hispanics in the USA (Fig. 7.4). Additional prob-
increment varies from 26 % (in the Amazonas) to lems are faced when they return to their coun-
153 % in Bogota [43]. Still a low prevalence has tries; it is well recognized that glycemic control
been found in rural communities. deteriorates, but, there is scant data on this topic.
In Argentina, the prevalence of T2D changed More studies are required to understand the inter-
from 8.4 to 9.6 % between 2005 and 2009. These action between genetic and environmental risk
numbers are based on patient self-report and a factors that contribute to the increased prevalence
nationwide survey. Argentina, Venezuela, and within these minority groups and between the
Uruguay have more than 90 % of their population ethnic groups living within their country and
living within urban areas. those living in the USA.

T2DM in Hispanics/Latinos Living

in the USA Gestational Diabetes
The Hispanic/Latino population comprises nearly
16 % of the total US population. They are the eth- The gestational diabetes screening programs
nic group with the highest prevalence of T2D applied in the region are under debate. Universal
(22.6 % (95CI % 18.427.5 % in 2012 NHANES)) screening is the recommended approach in
despite that they are not the group with the big- nearly 80 % of the reviewed guidelines. Although
gest mean body mass index (29.7 kg/m2, 95 % CI settings with weak health systems might struggle
29.230.1) [44, 45]. Also, they are the group in to implement a universal screening approach,
which the proportion of undiagnosed cases is selective screening would risk missing up a large
greater (49 %, 95 % CI 40.857.2 %) and the proportion of cases. Selection would be compli-
achievement of treatment goals is the lowest. cated given the nature of some of the most com-
Isolation, language barriers, and lack of access to mon risk factors used to determine the need for
7 Diabetes in Latin America 113

screening. For example, history of previous those with low Amerindian admixture. Temporal
GDM is likely to be missed. Likewise, ethnicity patterns and interaction with environmental fac-
as a risk factor would qualify all patients in tors were postulated as potential explanations for
African, Asian, and Latin American settings as the changes in incidence rates [48, 49].
high risk.
T1DM in Uruguay
The Uruguayan population is, demographically, a
Type 1 Diabetes mixture of Caucasian (Spain, Italy, France,
Portugal, and Lebanon), Negroid (Congo and
There is a lack of national registries and large Angola), and Amerindian (mainly Charras,
cohort studies in the region. Evidence is derived Minuanes, and Guaranies) groups; during the
mainly from single-center studies. Diagnoses and seventeenth and eighteenth centuries, their
treatment of T1D are limited to reference centers. groups were admixed, making it impossible now-
Ketoacidosis remains as a common initial T1D adays to find native unmixed ethnic groups as
manifestation. A Brazilian registry (20082010) opposed to most Latin American countries.
informed that T1D diagnosis was made based on Interestingly, 92 % of the population in
the presence of ketoacidosis in 42.3 %; this per- Montevideo is of Caucasian origin, representing
centage is greater than that reported in the USA one of the cities with the highest T1DM inci-
(29 %) or in some countries of Europe (<20 %). dence according to the DIAMOND study in Latin
Positive titers of anti-GAD antibodies were found America: 8.3 % (CI95 % 5.411.7) [50].
in 64 % of recently diagnosed Brazilian T1D
patients. There is no information regarding the T1DM in Mexico
number of at-risk individuals (based on the pres- Limited information regarding T1DM in Mexico
ence of positive titers of antibodies and/or an has been published. Data collected in the years
abnormal insulin secretion). 19781992 revealed a low incidence that was
A growing challenge is the differential diag- consistent with the figures issued by the
nosis of new-onset hyperglycemia in pediatric DIAMOND project group. The study was con-
populations. Nearly half of children with diabetes ducted under standardized methodology pro-
are obese and resistant to ketoacidosis and could posed by the WHO in an urban area at Boca del
be treated with oral glucose-lowering agents. Rio, a port located on the Gulf of Mexico in
between 1990 and 1993 [51]. The average inci-
T1DM in Chile dence rate was 1.15/100,000 per year (CI95 %
Between 1980 and 1993, a study conducted in a 0.751.9). These results have positioned Mexico
native aboriginal Mapuche population from Chile as the country with the lowest rates of T1D inci-
showed a very low incidence (0.43/100,000 per dence. Nevertheless, Gmez-Daz et al. [52]
year CI 95 % 00.95) of T1DM in children under recently reported remarkably greater incidence
the age of 14 [47]. These data were significantly rates using as source of documentation the regis-
different from that reported for the Caucasian tries of the Mexican Institute for Social Security
Chilean population in the same study (IMSS, in Spanish), the largest social healthcare
(1.58/100,000 CI 95 % 1.112.04; p < 0.0016). provider within the country. Incidence among
Even though Caucasian heritage came mainly children younger than 19 years of age was calcu-
from Spain, incidence of T1DM in this Chilean lated for a 10-year period (20002010); the num-
subgroup was seven times lower than that ber of new T1DM cases increased from 3.4 to 6.2
reported for Madrid or Catalonia. A significant per 100,000 insured cases during the study. The
increase in incidence was observed in the period highest incidence rate was observed in 2006,
20012004 (5.4 vs 8.33/100,000 inhabitants per with 1,029 new cases within a population of
year, p < 0.04). Higher rates were observed in 11,739,112 (8.8 new cases/100,000 insured pedi-
higher-income strata, urbanized counties, and atric subjects). The age groups with the biggest
114 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

increment in number of cases were between the lower-than-expected number of cases among
ages of 1014 years (2.1-fold increase between Mexicans, Blacks, and Oriental populations.
2003 and 2010) and 1519 years (1.9-fold Other comparative ethnic studies include the
increase between 2003 and 2010). This study SEARCH study, the Philadelphia Registry, and
revealed a substantial increasing trend in T1DM the Allegheny and Colorado IDDM studies [55].
incidence in Mexican children less than 19 years.
Even though causative factors such as perinatal
infections, weight gain in the first months of life, Complications of Diabetes
and increased maternal age might be related to in the Region
the outcome, it must be considered that inclusion
criteria were different for both studies. Information regarding diabetes-related chronic
complications is scant in LA. Screening proce-
T1DM in Argentina dures for retinopathy, nephropathy, and foot
Four Argentinean centers were included in the problems are performed without proper system-
DIAMOND project. Avellaneda was the most atization. In the 2012 ENSANUT survey, screen-
representative city of Caucasian population; data ing for albuminuria, retinopathy, and foot
were obtained from a retrospective study con- abnormalities were performed in 34.2 %, 38.6 %,
ducted from 1985 to 1988 (obtained from the and 59.3 % of the T2D cases, respectively, during
national census). In addition a prospective survey the previous year. In the same population-based
was conducted from 1988 to 1994; the primary survey, 14.6 % referred having some degree of
sources were preprimary and primary schools visual problems, 13.4 % had lost sensitivity in at
with secondary sources considering pediatric least one part of their bodies, 9.4 % reported hav-
hospitals, private diabetologists, and pediatri- ing had ulcers in legs or feet, 4.9 % were blind,
cians in Avellaneda, its surroundings, and Buenos 3 % had some amputation, 2.3 % had been diag-
Aires City. An annual incidence rate ranging nosed with a diabetic foot, and 1.2 % were on
from 6.89/100,000 per year (CI95 % 4.389.4) in dialysis [56].
1985 to 7.59/100,000 per year (CI95 % 6.41 T2D is the main cause of premature disability,
8.77) in 1990 was observed [53]. blindness, end-stage renal disease (ESRD), and
nontraumatic amputations as well as one of the
Hispanics Living in the USA ten most frequent cases of hospitalization in
Data regarding T1DM prevalence in Hispanics adults.
living within the USA is scarce. Orchard et al. ESRD is a major cause of premature disability
analyzed the NHANES 19992010 sample and premature mortality in the region. Registries
(n = 59,130). The population-based design of the are available in many LA countries, but in few
survey and the small number of cases limited the instances the contributions of diabetes and hyper-
ability of the authors to provide precise estimates tension (main causes of ESRD in all countries)
of the prevalence. The Hispanic population had are estimated separately [57]. The Latin American
half the prevalence of that reported in Caucasians Dialysis and Renal Transplant Registry, founded
and African Americans. The same conclusion in 1991, collected the available evidence in 2010.
was reached when the Mexican American sub- The ESRD prevalence has increased in the region
jects were analyzed as a separate group [54]. from 119 patients per million in 1991 to 660
The T1DM incidence has been compared in patients per million in 2010. The highest preva-
US samples between groups with different ethnic lence was reported in Puerto Rico (1,355 patients
backgrounds. One of the first studies reporting per million) followed by Mexico, Argentina,
incidence was published in 1985 and was con- Uruguay, and Chile (between 777 and 1,136
ducted in southern California between 1978 and patients per million). The higher incidence rates
1981; the observed incidence rates showed an were reported for Mexico (458 patients per mil-
excess of cases in the Caucasian population and a lion year), Puerto Rico, Argentina, Brazil, and
7 Diabetes in Latin America 115

Chile. The most common alternative to replace 0.92.2 % of the adults older than age 50 were
kidney function is hemodialysis (75 % of treated functionally blind. Diabetic retinopathy was
ESRD cases), but large differences exist between responsible of 19 % of the cases in which the
countries and health systems. The kidney trans- cause was recorded. Other common causes were
plant rate increased from 3.7 to 6.9 patients per age-related macular degeneration (26 %) and
million in the 19912010 period. Among coun- glaucoma (26 %). The contribution of diabetic
tries in which the ESRD cause is registered, the retinopathy varied between countries, being the
contribution of diabetes is the highest in Puerto highest (>30 %) in Argentina, Paraguay, and
Rico (66.8 %) and Mexico (61.8 %). The smallest Colombia. Furtado and coworkers collected the
contributions were found for Cuba (26.2 %) and evidence published until 2012 regarding the
Uruguay (23.2 %). prevalence of diabetic retinopathy in the LA
The ESRD incidence reported for Mexico is region. He found 63 studies from 11 countries.
the highest worldwide. This observation is in However, a large proportion of the reports are
accordance with the increased susceptibility for based in relatively small, biased populations
having microvascular complications found in (studied in reference centers).
Hispanics in the USA. This population has a Prevalence of diabetic retinopathy is higher in
threefold increased risk compared to Caucasians. Mexican Americans than in Caucasians, but this
Furthermore, the disease burden of ESRD in the difference was not statistically significant in the
population with diabetes is remarkably bigger in 20052008 NHANES survey. Despite that,
Mexico compared against reported in the Mexican Americans have a significantly higher
USA. The disability-adjusted life years (DALYs) risk for having blindness as a result of diabetic
rate due to diabetic kidney disease is several retinopathy (odds ratio 3.6 (95 % CI 1.0512.56)).
times greater in Mexico (103.1 vs 880.5 per There is scant information about the incidence of
100,000 inhabitants). The same is true for the diabetic retinopathy. The Mexico City Diabetes
annual change of DALYs rate (2.5 vs 6.5 % in the Study found a 3-year incidence of 23 % in 164
19902013 period) and the years of life lost cases [59].
(YLL) (836.2 vs 42.9 per 100,000 individuals). Lower-limb amputations remain as a major
In contrast, the years lost due to disability (YLD) cause of premature disability in the region.
are greater in the USA (44.3 vs 60.2 per 100,000 Diabetes is the cause of 70 % of nontraumatic
subjects). Thus, the distribution of the disease lower extremity amputations. Despite that,
burden is different between Mexico and the population-based information is scant. In the
USA. While in Mexico the burden is caused Mexican ENSANUT 2012 report, 2 % of the
mainly by premature mortality, premature dis- T2D participants had an amputation and 7.2 %
ability is the main contributor in the USA. had at least one ulcer in the lower limbs in the
There is a lack of information about the inci- past. In Costa Rica, the incidence of lower-
dence of diabetic nephropathy and the prevalence limb amputations was 6.02 per 1,000 patients
of the various stages of the disease using as per year among the population treated by a
source of evidence population-based surveys. social security system (20012007). This rate
The prevalence of diabetic retinopathy was is unacceptably high; it is similar to that
estimated at the LA region in 1999. Sixteen informed in other developing countries (13.7
countries and 7,715 patients participated. Any per 1,000 cases per year) and several times
kind of diabetic retinopathy was found in 42 % higher than that reported in Great Britain (2.8
of the cases; 17 % required immediate treatment. per 1,000 cases per year) and the USA (0.8 per
A large percentage had no previous eye exam 1,000 cases per year). Similar rates have
[58]. Other efforts in the region have use as out- informed in Brazil (4.3 %) and for Mexicans
come the functional loss of vision. Using infor- living in the US/Mexico border. In clinic-based
mation from 15 countries and 55,643 patients cohorts, the percentage of amputee subjects is
(20102013), Limburg and coworkers found that remarkably greater (13 % in a cohort from
116 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

Chile). In Mexico, 12 % of the hospital admis- The impact of diabetes in health systems
sions among T2D patients were due to lower- should be assessed following the Global
limb ulcers. Burden of Disease group (GBD) approach. It
Cardiovascular events are the major cause considers not only the contribution of the dis-
of mortality in the T2D population. The contri- ease to total mortality but also YLL (years of
bution of T2D among cases with an acute myo- life lost) and YLD (years of life with disability).
cardial infarction is greater than that found in Individual profiles of each one of the countries
Caucasians. In the RENASICA II registry (the of the region are available at the Web page of the
largest registry of acute coronary syndrome in GBD group (www.healthmetricsandevaluation.
LA, n = 4,555), 42 % of cases had T2D [60]. org). For Mexico, diabetes is the third cause of
This percentage is greater than that found in YLL due to premature death; this parameter
the ACCESS (33 %) and the INTERHEART grew 32 % in the time period 19902013. Also it
(7.2 %) studies. The same conclusion was pro- is the leading cause of DALY (with an incre-
posed by the authors of the REACH study [61] ment of 50 % from 1990 to 2013) and the fourth
(1,816 stable outpatients with established vas- cause of YLDs (with a growth of 102 % in the
cular disease). In this report, the prevalence of same time period).
T2D among cases with coronary artery disease, Brazil T2DM accounts for 5.1 % of disability-
cerebrovascular disease, and peripheral artery adjusted life years, according to the Global
disease was 37.2, 37.5, and 56.8 %, Burden of Disease project; when compared to
respectively. other countries, Brazil showed a higher propor-
Based on T2D patients data from the Mexican tion of life years with disability among total
ENSANUT 2006, Reynoso-Novern and disability-adjusted life years for T2DM. Registry
coworkers estimated that 112 cases per 1,000 data suggests that mortality standardized for age
persons with T2D will suffer at least one isch- and gender in people with diabetes was 57 %
emic coronary event within the next 20 years. In higher than that of the general population. The
the same period, there will be 889,433 new cases Brazilian Study on Diabetes Costs (ESCUDI)
of heart failure, 2,048,996 events of myocardial estimated the costs of diabetes care in the public
infarction, 798,188 stroke events, and 491,236 healthcare system by interviewing 1,000 patients
nontraumatic amputations attributable to and a retrospective analysis of their medical
T2D. The expected mortality rate is 539 per records; data suggests that 70.4 % of patients had
1,000 persons with T2D with an average life at least one microvascular complication, 17 %
expectancy of 10.9 years [62]. had at least one macrovascular complication, and
In less than 30 years (19702000), T2DM 16 % had at least one of both. ESCUDI estimated
moved from the fifteenth to the first position that the direct and indirect costs of T2DM were
among the main causes of death. This remark- 1,144 USD per patient at the primary care level,
able change is multifactorial. The coding sys- reaching 2,810 USD at the tertiary level. T2DM
tem of the death certificates was adjusted to is the fifth cause of death in Brazil. The control of
fully represent the contribution of chronic dis- hypertension, tobacco use, and dyslipidemia in
eases (instead of the attribution of death to the recent years has decreased cardiovascular mor-
final event). Recent data indicated that the mor- tality in T2DM patients from an 11 % increase in
tality rate for T2DM has progressively the 19962000 period to an 8 % reduction from
increased. In 2013, 89,420 of the 717,357 2000 to 2007.
reported deaths (12.46 %) occurred in T2D In Argentina, the crude diabetes mortality
patients. The diabetes mortality rate was 73.95 increased from 19.6 to 21.3 per 100,000 inhabit-
per 100,000 inhabitants. Mortality rates have ants. It is the seventh cause of death (7,701
had a larger increase for men; the average age deaths per year). The main cause of mortality is
at death was 66.7 years [63]. cardiovascular death. Chronic complications are
7 Diabetes in Latin America 117

common. Among patients with more than In a great proportion of the LA countries,
20 years of exposure, 2.6 % had ESRD, 15.7 % three independent sectors (public, social security,
had an amputation, and 6.9 % were blind. and private) provide health services. The public
Diabetes caused in 2005 1,328,802 DALYs; it is sector covers the population not attended by the
the ninth and 11th cause of YLL for women and two other providers. In many instances, the pub-
men, respectively. lic sector offers a basic package of services that
covers the ambulatory care of the main chronic
disease in primary care units at a low cost or free
Coordination and Delivery of charge. The packages usually include generic
of Diabetes Care Services forms of metformin, sulphonylureas, statins, ace-
tyl salicylic acid, and regular NPH insulin. In
Healthcare systems in Latin America have some countries, materials required for self-
evolved in synchrony with the socioeconomic glucose monitoring are provided. Limited or no
and political system [64]. At the end of the 1980s, access to specialists, certified laboratories, and
several countries in Latin America thrived for a reference centers and the lack of coverage for
reduction in poverty and income inequalities. ESRD or other major complications are common
Despite that, big disparities persist in the region deficiencies in this sector. The social security
and within countries. The gross domestic product sector is financed by a fixed contribution by the
(GDP) per capita varies from close to 4,000 USD employee, employers, and the government. A
(Honduras) to 21,000 USD (Chile). In the pro- large proportion of patients with chronic diseases
cess of those reforms, also, the organization and are treated at primary care units. Although
philosophy of their individual healthcare systems patients may have access to specialized services,
changed. Several Latin American healthcare sys- a major challenge is to avoid delays in giving
tems have implemented or are in the process to access to services. Finally, the private sector
implement healthcare reforms. Brazil, Cuba, and offers prepaid medical plans or independent med-
Costa Rica currently have unified healthcare sys- ical services.
tems with parallel subsystem organizations; other The Pan American Health Organization has
countries introduced a government-financed published several documents to help govern-
insurance scheme and healthcare provision to ments and organizations to improve the quality of
attempt a reduction at care inequalities. Despite diabetes care and to provide services to under-
that, four main barriers should be addressed: (1) served communities [66]. Recommendations are
the structural fragmentation of healthcare sys- based in the chronic care model using a patient-
tems, which reduces efforts of mitigating inequal- centered approach. With their support, several
ities in standards of care throughout the region; pilot studies have been implemented in Mexico
(2) the centralized decision-making processes; [67], Chile, and Bolivia with good results.
(3) the lack of regulation within the healthcare However, their impact has not been enough to
system, especially in terms of the quality of change local practices.
health-related services and quality of drugs in Several health systems have launched initia-
health systems; and (4) high costs and low effi- tives to provide specialized care at a low cost.
ciency. The percentage of the GDP allocated to In Mexico, the health ministry started the
cover health expenditures is below 10 % in the UNEMES program, which is a network of pri-
region (being Brazil the highest (9.7 %) and mary care units operated by a multidisciplinary
Venezuela the lowest (3.6 %)). A detailed analy- team (dietitian, psychologist, internist) that
sis of the health systems of the region is beyond applies a standardized intervention against
the scope of this manuscript. Interested readers T2D and other chronic diseases [68]. Some
may consult reference [65] for additional countries (e.g., Brazil and Mexico) have devel-
information. oped Internet-based monitoring systems of the
118 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

interventions (SysHiperDia) and policies large percentage of cases do not receive a sub-
focused in T2D [69]. In addition, reforms in stitutive therapy of renal function. Similar
the health insurance coverage have been imple- numbers have been informed several reports
mented to stimulate the adoption of a healthy from Argentina (reviewed in detail in Gonzalez
lifestyle among T2D patients (e.g., in et al. [73]).
Colombia) or to improve the performance of
health providers (e.g., in Chile) [70]. Some
conditional cash transfer programs have been Non-pharmacological Management
linked to preventive health programs.
The direct and indirect costs of T2D treat- More than 80 % of T2D cases are treated by pri-
ment are remarkable [71, 72]. In the region, the mary care physicians in the public and social
annual cost of diabetes healthcare is between security sectors. In many cases, there are sig-
34.6 and 59.9 billion US dollars; it represents nificant delays in getting an appointment, and
close to 12 % of the total healthcare budget. the time spent in each consultation is too short
This proportion is the same than mean percent- to provide proper care. A major problem of the
age worldwide. The average annual cost per region is lack of access to structured and effica-
person is between 1,169 and 2,027 US dollars, cious treatment programs. Providers are not
which is lower to that reported in the USA and properly trained and do not have stimulus to
Europe (5,3749,641 USD). There are large improve the quality of their services and their
differences in the amount of money spent in knowledge regarding T2D. Clinical inertia is a
diabetes healthcare costs between countries. In major challenge in the region. Many health sys-
2010, the biggest budgets were estimated for tems have dietitians involved in the treatment
Mexico (4,836 million USD) and Brazil (4,296 of T2D cases in primary care units, but, only a
million USD). few have multidisciplinary teams. Access to
In 2010, researchers from the National electronic medical records is limited to some
Institute of Public Health in Mexico calculated social security systems. Referral to an endocri-
that the greater direct costs correspond to med- nologist is usually late and limited to cases with
ications (133,143,734 USD), followed by com- chronic complications. There are not enough
plication costs (110,410,928 USD), consult/ endocrinologists in the region. For example,
diagnosis-related costs (59,734.448 USD), and Mexico has close to 800 endocrinologists that
hospitalization costs (39,937,331 USD). are insufficient to treat the 11.5 million cases.
Indirect costs are mainly due to permanent dis- Despite of the above, the services provided by
ability (409,205,846 USD), followed by costs specialists are underutilized due to lack of
due to premature mortality (19,623,029 USD) awareness of the public, their primary care phy-
and costs due to temporal disability (6,372,059 sicians, and the health systems. In many coun-
USD). Thus, more than half of the cost is due tries, diabetes educators are not included in the
to indirect costs; as a result, conclusions are structure of health services. Certified HbA1c
heavily dependent on the methodology applied measurements are not available for a large pro-
to estimate the indirect costs. Out-of-pocket portion of the primary care clinics. Screening
payments are a major challenge in the region. for chronic complications and other preventive
The average annual cost per person was 3,193 actions (e.g., retinopathy screening) are per-
USD; the cost is bigger for cases with chronic formed at lower than expected rates. Table 7.4
complications (2,749 vs 3550 USD). These shows the percentage of cases in Mexico
numbers may be underestimated. The cost of (nationwide) and Argentina (Qualidiab net-
covering diabetes-related ESRD is not prop- work) in which preventive actions are imple-
erly included in these estimations because a mented. Although the representativeness of
7 Diabetes in Latin America 119

Table 7.4 Implementation of preventive actions against diabetes-related chronic complications in Mexico and
Preventive action (percentage, 95 % Mexico 2012 (Ref. Argentina Qualidiab 2006
CI) Mexico 2006 [56]) (Ref. [73])
Four or more medical evaluations per 58.8 (58.559.2) 65.4 (64.966.0)
HbA1c testing (at least two times per 3.7 (3.63.8) 7.7 (7.38.2) 40
Blood pressure measurements 50.5 (50.250.8) 67.9 (67.468.3) 99
Plasma lipid measurements 27.3 (27.027.6) 79.2 (78.779.7) 61
Microalbuminuria detection 6.6 (6.56.7) 12.6 (11.913.3) 8
Retinopathy detection 12.3 (12.112.4) 8.6 (8.19.0) 45
Diabetic foot detection 9.4 (9.29.5) 14.7 (14.115.2) 55
Follow a dietary and exercise plan 3.7 3.63.7) 6.8 (6.57.4) 2.3

these reports is different, some deficiencies are population that has access to them is small due
shared. Retinopathy screening and foot exams to limited infrastructure or economic reasons.
are not scheduled as needed despite that patients
have frequent contacts with the medical units.
A very small proportion of patients have Rational Selection of Anti-diabetes
received and implemented a dietary/exercise Medications
plan. These deficiencies could be reverted in a
short period of time by quality assurance pro- Multiple diabetes guidelines are available in the
grams and training of the health providers. region. Every country has legal documents that
Some countries have organized support net- regulate local practices. In addition, most national
works coordinated by patients and health profes- diabetes societies have position documents,
sionals. In addition, educational Web pages and which contain the opinions of the local experts.
24-h phone lines are available. However, some of Furthermore, the Latin American Diabetes
these efforts are mainly focused in providing Association (ALAD) published a set of evidence-
information rather than modifying behaviors. based guidelines that take in consideration the
Their impact at community level has not been existing information and local resources. Also,
measured in a systematic manner. Telemedicine the most frequently cited international guidelines
is available in some countries to provide services are widely disseminated among primary care
to communities with limited access. It includes physicians. Despite of the above, a large propor-
medical consultation and screening for some tion of them do not have an in-depth knowledge
chronic complications (e.g., diabetic retinopathy, of any of these documents.
diabetic foot). In some countries (i.e., Argentina) More than 80 % of patients receive pharmaco-
legislation has been modified to give the right to logic treatment for hyperglycemia control. It is
patients to have continuous access to insulin and based mainly in oral glucose-lowering drugs
glucose strips. (generally metformin with or without sulphonyl-
Inequality is a major challenge of the region. ureas). The most recently introduced drugs
Diabetes care is not an exemption. Despite of (DPP-IV inhibitors, SGLT2 inhibitors, GLP1
the above, there are highly qualified reference agonists) are available in the Latin American
centers that provide state-of-the-art care of markets, but, not in the majority of the public or
their patients in the majority of the countries of social security systems. The proportion of
the LA region. However, the percentage of the insulin-treated patients (620 %) [74, 75] is
120 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

below the international standards (3050 %). (NPH + regular insulin), administered in two
Insulin is usually administered as a single bed- daily injections; only 9 % received three daily
time dose in combination with oral agents. Insulin insulin injections.
analogs are available in some social security sys- The lack of effectiveness of the diabetes man-
tems. There is no information regarding the pat- agement programs is a multifactorial phenome-
tern of use of insulin based on nationwide surveys non. Explanations are classified as related to the
or registries. In addition, close to 10 % of cases healthcare provider, the service organization, and
are treated with herbal medicine. the patient. Clinical inertia, the lack of decision-
As a result, the achievement of the treatment making tools, and the absence of competence-
goals is lower to that reported in other regions. based training programs are the most common
For example, in the 2012 Mexican ENSANUT barriers for the healthcare provider. With respect
survey, only 25 % of the patients had an HbA1c to the healthcare system, obstacles include atten-
below 7 % despite that the vast majority of them tion processes designed to treat acute conditions,
have more than two medical visits per year and insufficient access to multidisciplinary teams,
were treated with oral agents. Still, 50 % of and other basic services and overcrowded facili-
patients had an A1c above 9 %. According to the ties. Depression, alcoholism, physical limita-
Qualidiab network, 57 % of T2D cases have tions, economic problems, and lack of awareness
fasting plasma glucose above 140 mg/dl in and knowledge of the disease are the most com-
South America [76]. In Brazil, a nationwide mon barriers among patients.
cross-sectional study evaluating 5,750 T2DM
patients in between 2006 and 2011 found mean
HbA1c levels of 8.6 2.2 %, with a median of Translating Primary Prevention
8.1 % where only 48.5 % of patients had HbA1c of Type 2 Diabetes
levels <8 %.
Treatment of comorbidities share the same The region has undergone rapid socioeconomic
problems than that mentioned for the manage- changes moving from a predominantly rural soci-
ment of hyperglycemia. Statins or antiplatelet ety to a westernized lifestyle in less than half a
agents are prescribed in a lower than expected century. Migratory movements to urban centers
percentage of cases. In Mexico, according to the and to the more developed countries exist in a
ATP-III guidelines, 71 % of T2D patients qualify large proportion of the LA countries [79]. As a
for statin therapy; in contrast, less than 10 % of result, dramatic shifts in food availability, food
the cases were treated and controlled at the preferences, and physical activity have happened.
moment of the survey. In addition, nearly 50 % of LA countries, especially Mexico, are among the
T2D cases had high blood pressure; less than biggest consumers of soft drinks and caloric-
10 % are treated and controlled. Similar rates dense processed food. The proportion of the
have been found in Colombia. A large proportion Mexican adult population who does not meet the
of the hypertensive or dyslipidemic cases receive minimum WHO-advised physical activity has
drug therapy, but only a few achieve treatment increased by 6 % between 2006 and 2012. Only
targets. The achievement of treatment goals is 28.8 % of Mexican adults had more than 150 min
below to that reported in the USA [77, 78]. per week of moderate-intensity physical activity.
The Qualidiab network analyzed the efficacy As a result, the adoption of a healthy lifestyle for
of T1D treatment in patients living in several cit- at-risk subjects and T2D cases is a challenging
ies of Argentina and South America. A quarter of process. A growing trend in the region is to
them had a blood glucose level of <80 mg/dl, and reduce the duration of the lactation period due to
41 % had a glucose value >140 mg/dl. Only one- lack of adequate facilities at the work places. As
quarter of the patients could play an active, effec- a result, women do not return to the preconcep-
tive role in DM control and treatment. Half of tion weight leading to an increased risk for hav-
them were treated with a mixed dose of insulin ing T2D. In addition, there is a lack of
7 Diabetes in Latin America 121

T2D-preventive programs applicable for women 10-year period. For Brazil, the estimated cost of
with gestational diabetes. Finally, air pollution the integrated package of preventive actions is
and exposure to several environmental contami- 0.4 USD per year per capita per DALY. However,
nants (e.g., arsenic, lead) are highly prevalent in the individual cost of each public policy is
some areas of the region. These are known risk remarkably different and varies between coun-
factors for having T2D. Each one of these risk tries. A health in all policies approach and a
factors is a target for intervention to prevent coordinating body at the highest social and gov-
chronic diseases. ernmental level are critical to assure the success
Several countries have enforced action plans of the prevention plans.
to mitigate the impact of diabetes. The Pan National and regional (Asociacin
American Health Organization proposed a set of Latinoamricana de Diabetes) nongovernmental
potential actions for the region in accordance associations are active stakeholders in the region
with the recommendations by the World Health [85]. Several academic bodies have published
Organization. Taxes for sugary beverages and pro-action documents; some include critical anal-
caloric-dense industrialized products [80], mass yses of the current approaches and alternative
media campaigns, school-based interventions proposals [86].
[81], various forms of food labeling [82], work-
site interventions, food advertising regulation,
changes in urban settings, and public transporta- Organizing and Conducting
tion [83] are some examples. In some countries Diabetes Research in the Region
(e.g., Mexico, Brazil, Argentina, Guatemala, and
others), these proposals have become public poli- The remarkable impact of diabetes in the region
cies, and they have been integrated in a national has raised the attention of several international
plan against chronic diseases [84]. Still, no suffi- and regional agencies. For example, The
cient evidence has been published to assess the European-Latin American and the Caribbean
impact of these interventions. A PAHO group Health (EU-LAC Health) consortia, funded by
published the estimated impact based on a the European Union, have sponsored calls for
simulation model (the chronic disease prevention projects to develop low-threshold interventions
model) for Brazil, Mexico, and Canada. At the to tackle diabetes and other chronic diseases [87].
individual level, intervention should increase the The Global Alliance for Chronic Diseases, a net-
consumption of fruits and vegetables, decrease work funded by several countries, facilitates the
the fat intake, raise the time devoted to physical interaction between LA researchers and groups
activity, and decrease the mean body mass index, from developed countries to design and validate
cholesterol, and systolic blood pressure of the high-impact interventions in developing coun-
population. According to them, fiscal measures tries [88]. Some national funding agencies (i.e.,
are the public policy with the biggest impact CONACYT) have funded problem-specific calls
(nearly 80,000 life years gained per year). Mass to tackle diabetes-related issues. Although these
media campaigns and school-based interventions efforts are remarkable, an integrated approach is
resulted in half of the effect estimated for taxa- needed to have a significant impact in the short
tion. In Mexico, primary care counseling was the term.
most effective intervention to modify the number Several regional initiatives should be high-
of disability-adjusted life years (DALYs) in the lighted. The CARMEN network addresses health
middle term. Some of these policies may take determinants and health equity [89]. Their prod-
two decades or more to have a significant impact ucts include implementation and evaluation of
on DALYs. The intervention that requires more public policies, social mobilization, community-
time to become cost-effective is the school-based based interventions, epidemiological surveil-
lifestyle modification. The authors assume that lance of risk conditions, and preventive healthcare
the implementation cost will be recovered in a services. The Slim Initiative in Genomic
122 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

Medicine for the Americas (SIGMA) Type 2 Research is a critical component to plan and
Diabetes Consortium, a collaboration between implement policies against diabetes. Still, there
the Broad Institute, the University of Southern are several LA countries in which population-
California, and several Mexican institutions based, nationwide epidemiological information
funded by a private source, has identified ethnic- is not available. In addition, there is scant or no
specific risk variants that explain a large propor- information in the region regarding the preva-
tion of the genetic susceptibility for having T2D lence of T2D in pediatric or unserved populations
of the Amerindian communities [16]. Finally, (e.g., Native Americans or migrants). Additional
FunPrecal is a collaborative network between studies are needed about the epidemiology of
several countries of the region with Finnish and gestational diabetes, prediabetes, monogenic dia-
US centers to validate preventive and therapeutic betes, and the majority of chronic complications.
interventions [90]. No incidence data exists in the majority of coun-
Innovation is a common requirement in the tries for many of the diabetes-related outcomes.
call for proposals published in the region. As a National registries and translational research are
result, a growing number of groups are working at an early stage in the region. Few papers had
on nanotechnology, new multidisciplinary informed the use of services or the development
patient-centered treatment schemes [91], tele- of diabetes-related economic models.
medicine, cell-phone-based interventions or new Despite of the above, there are unique research
diagnostic devices, or population-specific prog- opportunities for the study of diabetes in Latin
nostic tests. America. The region has peculiarities that may
be used to generate new local and global knowl-
edge to the field. For example, the ethnic compo-
Future Directions: Unmet Needs, sition of many countries of the region contains a
Unanswered Questions, large proportion of Amerindian heritage, a group
and Unquestioned Answers that has not been sufficiently represented in the
genetic consortia. In addition, the study of the
In summary, three major challenges exist in the interaction between infectious diseases with dia-
region to mitigate the impact of T2D. First, a betes may offer new insights about the effects of
large proportion of the population has preceding diabetes-related abnormalities on the immune
conditions that increase their risk in the midterm. response. Environmental phenomena linked to
Second, half of the patients are undiagnosed, pre- T2D (e.g., migration, exposure to pollutants or
cluding the implementation of preventive actions caloric dense foods and beverages) are rapidly
against chronic complications. Third, the effec- evolving in the region. Public policies against
tiveness of the treatment programs is below the T2D could be tested in a shorter period of time
international standards, resulting in high expen- than in other areas due to the large number of
ditures without changing the rates of premature diabetes-related outcomes occurring in LA.
disability and mortality. The number of prevent- Actions to control the T2D outbreak in LA
able hospital admissions and deaths could be should come from the whole society. Although it
greatly diminished by the implementation of is critical to have the support of the authorities at
nationwide quality assurance programs. Budget the highest level, it is unlikely that governmental
allocation should be based on economic models plans could be sufficient to mitigate the health
and cost-effective interventions. Incentive pro- and economic consequences of the disease.
grams may be helpful to increase productivity Academic institutions, NGOs, and social leaders
and quality of services; those programs should be are crucial for keeping T2D among the top health
adapted to the local culture and needs in order to priorities. Also, they introduce the innovation and
create win-win relationships. A detailed analysis the scientific background of the action plans.
of the potential solutions is beyond the scope of Everybody could contribute. At the end, preven-
this manuscript; interested readers should consult tion and treatment of T2D heavily depend on per-
references [70, 86, 92]. sonal choices that determine peoples lifestyle.
7 Diabetes in Latin America 123

References 11. Chan JCN, Gagliardino JJ, Baik SH, Chantelot J-M,
Ferreira SRG, Hancu N, et al. Multifaceted determi-
nants for achieving glycemic control: the International
1. Saad PM. Demographic trends in Latin America and
Diabetes Management Practice Study (IDMPS).
the Caribbean. In: CEPAL 2010. http://www.cepal.
Diabetes Care. 2009;32:22733.
12. Collado-Mesa F, Barcel A, Arheart KL, Messiah
SE. An ecological analysis of childhood-onset type 1
ED_12_7_09.pdf. Accessed 26 Sept 2015.
diabetes incidence and prevalence in Latin America.
2. Danaei G, Finucane MM, Lu Y, Singh GM, Cowan
Rev Panam Salud Publica. 2004;15(6):38894.
MJ, Paciorek CJ, Lin JK, Farzadfar F, Khang YH,
13. Karvonen M, Viik-Kajander M, Moltchanova E,
Stevens GA, Rao M, Ali MK, Riley LM, Robinson
Libman I, LaPorte R, Tuomilehto J, Diabetes
CA, Ezzati M, Global Burden of Metabolic Risk
Mondiale (DIAMOND) Project Group. Incidence of
Factors of Chronic Diseases Collaborating Group
childhood type 1 diabetes worldwide. Diabetes Care.
(Blood Glucose). National, regional, and global trends
in fasting plasma glucose and diabetes prevalence
14. Stanescu DE, Lord K, Lipman T. The epidemiology
since 1980: systematic analysis of health examination
of type 1 diabetes in children. Endocrinol Metab Clin
surveys and epidemiological studies with 370
N Am. 2012;41(4):67994.
country-years and 27 million participants. Lancet.
15. The DIAMOND Project Group. Incidence and trends
of childhood type 1 diabetes worldwide 19901999.
3. Food and Agriculture Organization of the United
Diabet Med. 2006;23(8):85766.
Nations. The state of food and agriculture 2013. In:
16. The SIGMA Type 2 Diabetes Consortium. Sequence
FAO 2013.
variants in SLC16A11 are a common risk factor for
i3300e.pdf. Accessed 25 Sept 2015.
type 2 diabetes in Mexico. Nature. 2014;506:97101.
4. Aguilar-Salinas CA, Rojas R, Gmez-Prez FJ,
17. The SIGMA Type 2 Diabetes Consortium. Association
Garca E, Valles V, Ros-Torres JM, et al. Early
of a low-frequency variant in HNF1A with type 2 dia-
onset type 2 diabetes in a Mexican, population-
betes in a Latino population. JAMA.
based, nation-wide survey. Am J Med.
18. Lorenzi M, Cagliero E, Schmidt NJ. Racial differ-
5. Del Brutto OH, Mera RM, Atahualpa Project
ences in incidence of juvenile-onset type 1 diabetes:
Investigators. Indices of abdominal obesity may be
epidemiologic studies in Southern California.
better than the BMI to discriminate Latin American
Diabetologia. 1985;28(10):7348.
natives/mestizos with a poor cardiovascular status.
19. Vehik K, Hamman R, Lezotte D, Norris JM,
Diabetes Metab Syndr. 2014;8(2):1158.
Klingensmith G, Bloch C, Rewers M, Dabelea
6. Rojas R, Aguilar-Salinas CA, Jimenez A, Shamah T,
D. Increasing incidence of type 1 diabetes in 017
Rauda J, Avila-Burgos L, Villalpando S, Lazcano
year-old Colorado youth. Diabetes Care.
Ponce E. Metabolic syndrome in mexican adults.
Results from the national health and nutrition survey
20. Pettitt DJ, Talton J, Dabelea D, Divers J, Imperatore
2006. Salud Publica Mex. 2010;52(supl1):S118.
G, Lawrence JM, Liese AD, Linder B, Mayer-Davis
7. Gagliardino JJ, de la Hera M, Siri F. Evaluation of the
EJ, Pihoker C, Saydah SH, Standiford DA, Hamman
quality of care for diabetic patients in Latin America.
RF, SEARCH for Diabetes in Youth Study Group.
Rev Panam Salud Publica. 2001;10(5):30917.
Prevalence of diabetes mellitus in U.S. youth in 2009:
8. Aguilar-Salinas CA, Velazquez Monroy O, Gmez-
the SEARCH for diabetes in youth study. Diabetes
Prez FJ, Gonzalez Chvez A, Lara Esqueda A,
Care. 2014;37:4028.
Molina Cuevas V, Rull-Rodrigo J, Tapia Conyer R,
21. Aschner P. Diabetes trends in Latin America. Diabetes
ENSA 2000 Group. Characteristics of the patients
Metab Res Rev. 2002;18 Suppl 3:S2731.
with type 2 diabetes in Mxico: results from a large
22. Atun R, Odorico-Monteiro de Andrade L, Almeida G,
population-based, nation-wide survey. Diabetes Care.
Cotlear D, Dmytraczenko T, Frenz P, Garcia P,
Gmez-Dants O, Knaul F, Muntaner C, Braga de
9. Mehta R, Del Moral ME, Aguilar Salinas
Paula J, Rgoli F, Castell-Florit P, Wagstaf A. Health-
CA. Epidemiologia de la diabetes en el anciano. Rev
system reform and universal health coverage in Latin
Invest Clin. 2010;62:30511.
America. Lancet. 2015;385:123047.
10. Delgado-Snchez G, Garca-Garca L, Castellanos-
23. Whiting D, Guariguata L, Weil C. IDF diabetes atlas:
Joya M, Cruz-Hervert P, Ferreyra-Reyes L, Ferreira-
global estimates of the prevalence of diabetes for
Guerrero E, Hernndez A, Ortega-Baeza VM,
2011 and 2030. Diabetes Res Clin Pract.
Montero-Campos R, Sulca JA, Martnez-Olivares
ML, Mongua-Rodrguez N, Baez-Saldaa R,
24. Wild S, Roglic G, Green A, Sicree R, King H. Global
Gonzlez-Roldn JF, Lpez-Gatell H, Ponce-de-
prevalence of diabetes: estimates for the year 2000
Len A, Sifuentes-Osornio J, Jimnez-Corona
and projections for 2030. Diabetes Care.
ME. Association of pulmonary tuberculosis and dia-
betes in Mexico: analysis of the national tuberculo-
25. International Diabetes Federation. South and Central
sis registry 20002012. PLoS One.
America Region. In: IDF Atlas. 2015. http://www.idf.
124 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

org/regions/south-central-america. Accessed 12 Dec 38. Mendes ABV, Fittipaldi JAS, Neves RCS, et al.
2015. Prevalence and correlates of inadequate glycemic
26. Barcelo A, Rajpathak S. Incidence and prevalence of control: results from nationwide survey in 6,671
diabetes mellitus in the Americas. Pan Am J Publ adults with diabetes in Brazil. Acta Diabetol.
Health. 2001;10(5):3008. 2010;47:13745.
27. ALAD Guidelines 2013. In: ALAD Latinoamerica. 39. Viana LV, Leito CB, Kramer CK, et al. Poor glycae-
2013. Accessed mic control in Brazilian patients with type 2 diabetes
26 Sept 2015. attending the public healthcare system: a cross-
28. Serrano R, Garca-Soidn FJ, Daz-Redondo A, sectional study. BMJ Open. 2013;3:e003336.
Artola S, Franch J, Dez J, et al. Estudio de cohortes 40. Aschner P, Aguilar-Salinas CA, Aguirre L, Franco L,
en atencin primaria sobre la evolucin de sujetos con Gagliardino JJ, Seclen S, Vinocour M, on behalf of
prediabetes (PREDAPS). Fundamentos y the IDF Diabetes Atlas. Diabetes in South and Central
metodologa. Rev Esp Salud Publica. America: an update. Diabetes Res Clin Pract.
2013;87:12135. 2014;103:23843.
29. Seclen SN, Rosas ME, Arias AJ, et al. Prevalence of 41. The Central America Diabetes Initiative (CAMDI):
diabetes and impaired fasting glucose in Peru: report survey of diabetes, hypertension and chronic disease
from PERUDIAB, a national urban population based risk factors. Belize, San Jos, San Salvador, Guatemala
longitudinal study. BMJ Open Diabetes Res Care. City, Managua and Tegucigalpa PAHO. 2011. http://
2015;3:e000110. doi:10.1136/bmjdrc-2015-000110. w w w . p a h o . o r g / h q / i n d e x .
30. Nucci LB, Toscano CM, Maia ALM, Fonseca CD, php?gid=16710&option=com_docman&task=doc_
Britto MMB, Duncan BB, Schmidt MI, Brazilian view. Accessed 12 Oct 2015.
National Campaign for Diabetes Mellitus Detection 42. Schargrodsky H, Hernndez-Hernndez R,
Working Group. A nationwide population screening Champagne BM, Silva H, Vinueza R, Silva Ayaguer
program for diabetes in Brazil. Rev Panam Salud LC, Touboul PJ, Boissonnet CP, Escobedo J, Pellegrini
Publica. 2004;16(5):3207. F, Macchia A, Wilson E, CARMELA Study
31. Gonzlez-Villalpando C, Dvila-Cervantes CA, Investigators. CARMELA: assessment of cardiovas-
Zamora-Macorra M, Trejo-Valdivia B, Gonzlez- cular risk in seven Latin American cities. Am J Med.
Villalpando ME. Risk factors associated to diabetes in 2008;121(1):5865.
Mexican population and phenotype of the individuals 43. Aschner P. Edemiologia de la diabetes en Colombia.
who will convert to diabetes. Salud Publica Mex. Av Diabetol. 2010;26:95100.
2014;56:31722. 44. Schneiderman N, Llabre M, Cowie CC, Barnhart J,
32. Meza R, Barrientos-Gutierrez T, Rojas-Martinez R, Carnethon M, Gallo LC, Giachello AL, Heiss G,
Reynoso-Novern N, Palacio-Mejia LS, Lazcano- Kaplan RC, LaVange LM, Teng Y, Villa-Caballero
Ponce E, Hernndez-vila M. Burden of type 2 dia- L, Avils-Santa ML. Prevalence of diabetes among
betes in Mexico: past, current and future prevalence Hispanics/Latinos from diverse backgrounds:
and incidence rates. Prev Med. 2015;81:44550. the Hispanic Community Health Study/Study of
33. Olaiz-Fernndez G, Rojas R, Aguilar-Salinas C, Latinos (HCHS/SOL). Diabetes Care.
Rauda J, Villalpando S. Diabetes mellitus in Mexican 2014;37(8):22339.
adults: results from the 2000 National Health Survey. 45. Heiss G, Snyder ML, Teng Y, Schneiderman N,
Salud Publica Mex. 2007;49:3317. Llabre MM, Cowie C, Carnethon M, Kaplan R,
34. Villalpando S, Rojas R, Shamah-Levy T, Avila MA, Giachello A, Gallo L, Loehr L, Avils-Santa
Gaona B, De la Cruz V, et al. Prevalence and distribu- L. Prevalence of metabolic syndrome among
tion of type 2 diabetes mellitus in Mexican adult pop- Hispanics/Latinos of diverse background: the
ulation. A probabilistic survey. Salud Publica Mex. Hispanic Community Health Study/Study of Latinos.
2010;52 Suppl 1:S1926. Diabetes Care. 2014;37(8):23919.
35. Gutierrez JP, Dommarco J, Shamah T, Villalpando S, 46. Cusi K, Ocampo GL. Unmet needs in Hispanic/Latino
Franco A, Cuevas L, Romero M, Hernndez patients with type 2 diabetes mellitus. Am J Med.
M. Encuesta Nacional de Salud y Nutricin 2012. 2011;124 Suppl 10:S29.
Resultados Nacionales. In: Instituto Nacional de 47. Carrasco E, Prez-Bravo F, Dorman J, Mondragn A,
Salud Pblica. 2012. Santos JL. Increasing incidence of type 1 diabetes in a
ENSANUT2012ResultadosNacionales.pdf. Accessed population from Santiago of Chile: trends in a period
11 Oct 2015. of 18 years (19862003). Diabetes Metab Res Rev.
36. Burke JP, Williams K, Haffner SM, Villalpando CG, 2006;22(1):347.
Stern MP. Elevated incidence of type 2 diabetes in San 48. Carrasco E, ngel B, Codner E, Garca D, Ugarte F,
Antonio, Texas, compared with that of Mexico City. Bruzzone ME, Prez F. Type 1 diabetes mellitus inci-
Mex Diabetes Care. 2001;24(9):15738. dence in Santiago, Chile. Analysis by counties in the
37. Almeida-Pititto, Loureno Dias M, Franco de Moraes period 20002004. Rev Med Chile.
A, Ferreira A, Reis-Franco D, Goldberg-Eliaschewitz 2006;134(10):125864.
F. Type 2 diabetes in Brazil: epidemiology and man- 49. Gonzlez RN, Torres-Avils F, Carrasco E, Salas F,
agement. Diabetes Metab Syndr Obes Targets Ther. Prez F. Association of the incidence of type 1 diabe-
2015;8:1728. tes mellitus with environmental factors in Chile
7 Diabetes in Latin America 125

during the period 20002007. Rev Med Chile. American population: the REACH registry. Clin
2013;141(5):595601. Cardiol. 2012;35(8):4517.
50. Mimbacas A, Prez-Bravo F, Hidalgo PC, Javiel G, 62. Reynoso-Novern N, Mehta R, Almeda-Valdes P,
Pisciottano C, Grignola R, Jorge AM, Gallino JP, Rojas-Martinez R, Villalpando S, Hernndez-vila M,
Gasagoite J, Cardoso H. Association between diabe- Aguilar-Salinas CA. Estimated incidence of cardiovas-
tes type 1 and DQB1 alleles in a case control study cular complications related to type 2 diabetes in Mexico
conducted in Montevideo. Uruguay Genet Mol Res. using the UKPDS outcome model and a population-
2003;2(1):2935. based survey. Cardiovasc Diabetol. 2011;10(1):1.
51. Aude Rueda O, Libman IM, Altamirano N, Robles C, 63. Barquera S, Campos-Nonato I, Aguilar-Salinas C,
LaPorte RE. Low incidence of IDDM in children of Lopez-Ridaura R, Arredondo A, Rivera-Dommarco
Veracruz-Boca del Rio, Veracruz. Diabetes Care. J. Diabetes in Mexico: cost and management of diabe-
1998;21(8):13723. tes and its complications and challenges for health
52. Gmez-Daz RA, Prez-Prez G, Hernndez-Cuesta policy. Glob Health. 2013;9:3.
IT, Rodrguez-Garca Jdel C, Guerrero-Lpez R, 64. Strengthening health systems: the role and promise of
Aguilar-Salinas CA, Wacher NH. Incidence of type 1 policy and systems research. Geneva, Alliance for
diabetes in Mexico: data from an institutional register Health Policy and Systems Research, 2004. Available
20002010. Diabetes Care. 2012;35(11):e77. at:
53. Sereday M, Mart M, Damiano M, Moser Strengthening_complet.pdf.
M. Establishment of a registry and incidence of 65. Atun R, de Andrade LO, Almeida G, Cotlear D,
IDDM in Avellaneda, Argentina. Diabetes Care. Dmytraczenko T, Frenz P, Garcia P, Gmez-Dants O,
1994;17(9):10225. Knaul FM, Muntaner C, de Paula JB, Rgoli F, Serrate
54. Menke A, Orchard TJ, Imperatore G, Bullard KM, PC, Wagstaff A. Health-system reform and universal
Mayer-Davis E, Cowie CC. The prevalence of type 1 health coverage in Latin America. Lancet.
diabetes in the United States. Epidemiology. 2015;385(9974):123047.
2013;24(5):7734. 66. Barcelo A, Luciani S, Agurto I, Orduez P, Tasca R,
55. Orchard TJ, Secrest AM, Miller GR, Costacou T. In Sued O. Improving chronic illness care through inte-
the absence of renal disease, 20 year mortality risk in grated health services delivery networks. Washington,
type 1 diabetes is comparable to that of the general DC: PAHO HQ Library; 2012. PAHO.
population: a report from the Pittsburgh epidemiology 67. Barcel A, Cafiero E, de Boer M, Mesa AE, Lopez
of diabetes complications study. Diabetologia. MG, Jimnez RA, Esqueda AL, Martinez JA, Holguin
2012;53(11):23129. EM, Meiners M, Bonfil GM, Ramirez SN, Flores EP,
56. Flores-Hernndez S, Saturno-Hernndez PJ, Reyes- Robles S. Using collaborative learning to improve
Morales H, Barrientos-Gutirrez T, Villalpando S, diabetes care and outcomes: the VIDA project. Prim
Hernndez-vila M. Quality of diabetes care: the Care Diabetes. 2010;4(3):14553.
challenges of an increasing epidemic in Mexico. 68. Crdova-Villalobos JA, Barriguete-Melndez JA,
Results from two national health surveys (2006 and Lara-Esqueda A, Barquera S, Rosas-Peralta M,
2012). PLoS One. 2015;10(7):e0133958. Hernndez-Avila M, de Len-May ME, Aguilar-
57. Machado-Alba JE, Moncada-Escobar JC, Gaviria Salinas CA. Chronic non-communicable diseases in
H. Quality and effectiveness of diabetes care for a Mexico: epidemiologic synopsis and integral pre-
group of patients in Colombia. Rev Panam Salud vention. Salud Publica Mex. 2008;50(5):41927.
Publica. 2009;26(6):52935. 69. Accessed Nov 16 2015.
58. Villena JE, Yoshiyama CA, Snchez JE, Hilario NL, 70. Glassman A, Gaziano TA, Bouillon Buendia CP,
Merin LM. Prevalence of diabetic retinopathy in Guanais de Aguiar FC. Confronting the chronic dis-
Peruvian patients with type 2 diabetes: results of a ease burden in Latin America and the Caribbean.
hospital-based retinal telescreening program. Rev Health Aff (Millwood). 2010;29(12):21428.
Panam Salud Publica. 2011;30(5):40814. 71. Arredondo A, De Icaza E. The cost of diabetes in
59. Gonzlez-Villalpando C, Gonzlez-Villalpando ME, Latin America: evidence from Mexico. Value Health.
Rivera Martnez D, Stern MP. Incidence and pro- 2011;14(5 Suppl 1):S858.
gression of diabetic retinopathy in low income popu- 72. Seuring T, Archangelidi O, Suhrcke M. The economic
lation of Mexico City. Rev Invest Clin. costs of type 2 diabetes: a global systematic review.
1999;51(3):14150. Pharmacoeconomics. 2015;33(8):81131.
60. Jurez-Herrera , Jerjes-Snchez C. Risk factors, 73. Gonzalez L, Caporale JE, Elgart JF, Gagliardino
therapeutic approaches, and in-hospital outcomes in JJ. The burden of diabetes in Argentina. Glob J Health
Mexicans with ST-elevation acute myocardial infarc- Sci. 2014;7(3):12433.
tion: the RENASICA II multicenter registry. Clin 74. Lopez Stewart G, Tambascia M, Rosas Guzmn J,
Cardiol. 2013;36(5):2418. Etchegoyen F, Ortega Carrin J, Artemenko S. Control
61. Cant-Brito C, Chiquete E, Ruiz-Sandoval JL, of type 2 diabetes mellitus among general practitio-
Gaxiola E, Albuquerque DC, Corbaln R, Ramos A, ners in nine countries of Latin America. Rev Panam
Bhatt DL, Steg PG, REACH Latin America Salud Publica. 2007;22(1):1220.
Collaborative Group. Atherothrombotic disease, tra- 75. Legetic B, Cecchini M. Applying modeling to
ditional risk factors, and 4-year mortality in a Latin improve health and economic policy decisions in the
126 O.Y. Bello-Chavolla and C.A. Aguilar-Salinas

Americas: the case of noncommunicable diseases. characteristics in a community based pediatric weight
Washington, DC: PAHO; 2015. control intervention. Int J Behav Nutr Phys Act.
76. Commendatore V, Dieuzeide G, Faingold C, Fuente 2014;11:17.
G, Lujan D, Aschner P, Lapertosa S, Villena J, Elgart 82. Hamlin R. Front of pack nutrition labelling, nutrition,
J, Gagliardino JJ, DIFAR Academic Committee. quality and consumer choices. Curr Nutr Rep.
Registry of people with diabetes in three Latin 2015;4:3239.
American countries: a suitable approach to evaluate 83. Becerra JM, Reis RS, Frank LD, et al. Transport and
the quality of health care provided to people with type health: a look at three Latin American cities. Cad
2 diabetes. Int J Clin Pract. 2013;67(12):12616. Saude Publica. 2013;29(4):65466.
77. Gakidou E, Mallinger L, Abbott-Klafter J, Guerrero 84. Available at Accesssed 16
R, Villalpando S, Ridaura RL, Aekplakorn W, Nov 2015.
Naghavi M, Lim S, Lozano R, Murray CJ. Management 85. Available at
of diabetes and associated cardiovascular risk factors 86. Available at
in seven countries: a comparison of data from national CAnivANM150/L15-Acciones-para-enfrentar-a-la-
health examination surveys. Bull World Health Organ. diabetes.pdf.
2011;89(3):17283. 87. Available at
78. Mendivil CO, Mrquez-Rodrguez E, Angel ID, Paz 88. Available at
G, Rodrguez C, Almada J, Szyskowsky 89. Available at
O. Comparative effectiveness of vildagliptin in com- 90. Available at:
bination with other oral anti-diabetes agents in usual- 91. Hernandez-Jimenez S, Garcia-Ulloa C, Mehta R,
care conditions: the EDGE-Latin America study. Curr Aguilar-Salinas CA, Kershenobich-Stalnikowitz
Med Res Opin. 2014;30(9):176976. D. Innovative models for the empowerment of patients
79. Reich D, Patterson N, Campbell D, Tandon A, with type 2 diabetes: the CAIPaDi program. Recent
Mazieres S, Ray N, Parra MV, Rojas W, Duque C, Pat Endocr Metab Immune Drug Discov.
Mesa N, Garca L, Triana O, Blair S, Maestre A, Dib 2014;8(3):2029.
J, Bravi C, Bailliet G, Corach D, Hnemeier T, 92. Tricco AC, Ivers NM, Grimshaw JM, Moher D,
Bortolini M, Salzano FM, Petzl-Erler M, Acua- Turner L, Galipeau J, Halperin I, Vachon B, Ramsay
Alonzo V, Aguilar-Salinas CA, Canizales-Quinteros T, Manns B, Tonelli M, Shojania K. Effectiveness of
S. Reconstructing native American population his- quality improvement strategies on the management of
tory. Nature. 2012;488:3704. diabetes: a systematic review and meta-analysis.
80. Sarlio-Lhteenkorva S, Winkler J. Could a sugar tax Lancet. 2012;379(9833):225261.
help combat obesity? BMJ. 2015;351:h4047. 93. Villalpando S, Shamah-Levy T, Rojas R, Aguilar-
doi:10.1136/bmj.h4047. Salinas CA. Trends for type 2 diabetes and other car-
81. Jelalian E, Foster G, Sato A, Berlin K, McDermott C, diovascular risk factors in Mexico from 19932006.
Sundal D. Treatment adherence and facilitator Salud Pblica Mx. 2010;52(supl1):S729.
Diabetes in the Caribbean
Michael S. Boyne

Introduction ethnic populations. There are English-speaking

(Anglophone) as well as Spanish-, Dutch- and
The Caribbean, also referred to historically as the French-speaking countries (Table 8.1), and there
West Indies, consists of the Caribbean Sea, its are differences in their cultures also. The majority
islands and the surrounding coasts. This large of Caribbean people have West African ancestry
geopolitical area is bordered by the Gulf of as a remnant of the African slave trade and there-
Mexico to the northwest, by the Northern Atlantic fore self-identify as black. Mixed-race and
Ocean to the northeast and by the coastline of European peoples of Dutch, English, French,
South America in the south. There are over 700 Italian and Portuguese ancestry make up a minor-
islands, islets and cays stretching over 4000 km ity. Genetic admixture studies showed rates of
containing a large melting pot of nations and cul- 1015 % for non-African ancestry [1]. Politically,
tures. The climate is mostly tropical, although it most of the Anglophone countries have formed a
is subtropical in the north, but is undergoing sig- socio-economic bloc called the Caribbean
nificant climate change at present. Community (CARICOM). There is moderate
Prior to European immigration in the late geographical mobility, and so several millions live
1400s, the population was estimated to be nearly outside of the Caribbean, mostly in the USA,
a million indigenous people (e.g. the Kalinago, Canada and the UK.
also known as Caribs, and the Arawaks such as The total regional population was 41.9 million
the Tano). These Amerindians were wiped out by in 2014. According to the World Bank, the
disease and genocide although a few remain in the median per capita income was $8995 in 2014
Eastern Caribbean. The influence of the European (, but the
powers, along with the importation of Africans as range is wide as countries vary from low to high
slaves, followed by Chinese and Indian inden- income (Table 8.1). Tourism is a major source of
tured labourers, as well as Middle Eastern mer- income for many Caribbean nations as the region
chants, has ensured that the Caribbean has diverse has a high literacy rate, tropical weather and rela-
tive political stability.
Life expectancy has been increasing over sev-
eral decades. It increased by 5 years between
M.S. Boyne, MD, FRCPC 1990 and 2013 [2], and it is now 74 years for men
Tropical Metabolism Research Unit, Tropical
and 77 years for women according to the Pan
Medicine Research Institute, The University of the
West Indies, Mona, Kingston 7, Jamaica American Health Organisation (PAHO) (http://
e-mail: Hence, the Caribbean

Springer International Publishing Switzerland 2017 127

S. Dagogo-Jack (ed.), Diabetes Mellitus in Developing Countries and Underserved Communities,
DOI 10.1007/978-3-319-41559-8_8
128 M.S. Boyne

Table 8.1 Populations of Caribbean countries, some of their developmental indicators, and age-adjusted diabetes
prevalence among adults age 2079 years in 2013
Life expectancy at GNI per capita in Age-adjusted
Country Income level Population (2014) birth (years) 2014 (USD) prevalence (%)
Dutch speaking
Aruba HI 103,400 75 24,990 13.6
Curacao HI 155,900 77 14.5
Sint Maarten HI 37,660 76 14.2
Suriname UMI 538,200 71 9470 11.1
English speaking
Anguilla 14,100 12.6
Antigua and HI 90,900 76 13,360 13.3
Bahamas HI 383,100 75 20,980 14.2
Barbados HI 283,400 75 14,960 12.4
Belize UMI 351,700 74 4350 15.9
Bermuda HI 65,180 81 106,140 12.8
British Virgin 28,100 12.6
Cayman Islands HI 59,170 14.3
Dominica UMI 72,340 77 7070 10.9
Grenada UMI 106,300 73 7850 9.4
Guyana LMI 763,900 66 4170 15.9
Jamaica UMI 2.721 million 73 5220 10.4
St Kitts and Nevis HI 54,940 71 14,490 13.0
St Lucia UMI 183,600 75 7080 8.2
St Vincent and UMI 109,400 73 6560 10.0
Trinidad and HI 1.354 million 70 15,550 13.0
Turks and Caicos HI 33,740
US Virgin Islands HI 104,200 80 13,660 12.1
French speaking
Haiti LI 10.57 million 63 820 6.7
Martinique HI 386,500 14.3
Guadeloupe HI 403,750 79 6.3
St Martin HI 31,530 79
Spanish speaking
Cuba UMI 11.38 million 79 5880 8.1
Dominican UMI 10.41 million 73 6030 11.4
Puerto Rico HI 3.548 million 79 19,310 13.0
1. Data obtained from the World Bank ( accessed on Nov 29, 2015)
2. Gross national income (GNI) per capita was calculated by the World Bank Atlas method converted to current US
dollars, divided by the midyear population
3. Diabetes prevalence rates from the International Diabetes Federation ( accessed on
Nov 27, 2015) that were estimated by logistic regression models to produce smoothed age-specific prevalence rates
for adults aged 2079 years for 2013. This allows for direct country-to-country comparisons
4. Income level can be high income (HI), upper-middle income (UMI), lower-middle income (LMI) or low income (LI)
8 Diabetes in the Caribbean 129

is rapidly ageing region. Simultaneously, infant expected seroprevalence of HTLV-I in Jamaicans

and child mortality has fallen. with type 1 diabetes (i.e. 17 %). However, this ini-
Coupled with these demographic changes, the tial finding is insufficient to prove a causal associa-
Caribbean is undergoing a rapid nutritional and tion and needs further study. Interestingly, HTLV-I
epidemiologic transition like many regions of the virions use the glucose transporter type 1 (GLUT1)
world [3]. In the 1940s, the most common causes to infect CD4(+) lymphocytes [9].
of morbidity and mortality were infections.
However, over the past 40 years, communicable
diseases account for less than 10 % of the total Type 2 Diabetes
mortality. Atherosclerotic complications, i.e. cor-
onary artery disease and stroke, are now the lead- Role of Lifestyle Factors
ing cause of mortality (accounting for ~7 out of Obesity is rising in the Caribbean as part of the
10 deaths), and diabetes is the third reported global pandemic. The evidence for its role as a
cause in several countries. At the same time, sev- pathophysiologic driver of incident diabetes is
eral Caribbean countries have a double burden of overwhelming, and this is true also for the
infectious diseases (especially HIV/AIDS and Caribbean. The population-attributable risk of
newly emerging diseases such as chikungunya) body mass index for incident diabetes in the
and non-communicable chronic diseases (NCDs), Caribbean is 66 %, and it is 80 % for waist/hip
which are assuming epidemic proportions con- ratio, highlighting the role of adiposity [10]. The
tributing to morbidity and mortality [4]. International Collaborative Study on
Hypertension in Blacks (ICSHIB), which was an
ecological study on the burden of NCDs in the
Classication and Unique Aspects African Diaspora, showed that the prevalence of
of the Pathophysiology of Type 1 obesity increased progressively from West Africa
and Type 2 Diabetes in the Region to the Caribbean and then to North America [11
13]. This geographical gradient in obesity also
Type 1A parallels the gradient in the per capita gross
national product supporting a nutritional transi-
Exposure to human enteroviruses has been impli- tion as the cause. Also worrisome is that the rates
cated as an environmental factor that could trig- of annual weight increase in Jamaicans (1.37 kg/
ger and accelerate the autoimmunity leading to year) are significantly greater than African
type 1A diabetes. In Cuban studies, the presence Americans (0.52 kg/year) and Nigerians (0.31 kg/
of enterovirus RNA in sera from newly diag- year) [14]. Presumably this steep rise in weight
nosed patients with type 1 diabetes was signifi- gain is due to the effects of rapid cultural changes
cantly higher than in healthy controls (27 % vs. in a transitional society, and it does not bode well
3 %) [5, 6]. Also, enterovirus RNA was detected for incident diabetes rates.
in sera of first-degree relatives of islet cell Caribbean women have a disproportionate
antibody-positive patients compared to healthy burden of obesity, as approximately one third of
controls (16 % vs. 0 %). Molecular mimicry or Caribbean women were obese and another third
the hygiene hypothesis are plausible explanations were overweight, and these rates are increasing in
for these phenomena similar to other populations many countries, such as Barbados and Dominica
and regions. [11]. Obesity is about half as common in men.
The human T lymphotropic virus type 1 This sexual dimorphism for obesity is different to
(HTLV-I) virus is endemic in the Caribbean with a other regions of the world where there is parity,
prevalence of 25 % [7]. While endocrine disorders or even greater risk, among men.
such as hypercalcaemia have been reported with The prevalence of diabetes is closely corre-
HTLV-I, there is relatively little data about hyper- lated to the BMI and the amount of intra-abdominal
glycaemia. One study [8] found a higher than fat (as measured by the waist circumference)
130 M.S. Boyne

across the east-to-west gradient of the Diaspora rise in diabetes mortality followed 6 years later.
(Figs. 8.1 and 8.2) [15]. The trends in diabetes However, a notable exception is that Indo-
incidence and mortality paralleled secular Trinidadian men do not have strong correlation of
changes in obesity. As evidence, obesity rates in diabetes prevalence with BMI [17], although this
Cuba plummeted in the early 1990s driven by an may reflect differences in intra-abdominal adi-
economic crisis, but there was a rebound fol- pose tissue deposits.
lowed by an overshoot several years later to its There is a sexual dimorphism in diabetes prev-
current prevalence of ~53 % when the economy alence as women have higher rates [18]. Although
and nutrition improved [16]. The population- much of the variance is due to the higher preva-
wide increase in weight was immediately fol- lence of obesity in women, there is significant
lowed by a 116 % increase in diabetes prevalence residual confounding by other factors, and this
and 140 % increase in diabetes incidence. A 49 % area needs more study. Clinical prediction mod-
els for diabetes in the Caribbean rely on the pre-
dictive ability of obesity using cut points of BMI
(30 kg/m2) or central obesity (waist circumfer-
Prevalence of diabetes

UK US ence >94 cm in men and 80 cm in women) [19,

10 UK 20]. However, waist circumference is not supe-
caribbean rior to BMI in Jamaicans in its ability to predict
incident diabetes [21]. Other data however sug-
6 caribbean Men gest that the BMI cut-offs should be lower, i.e.
4 Women
24.8 kg/m2 (men) and 29.3 kg/m2 (women). For
West waist circumference, these would be 88 cm and
0 84.5 cm for men and women, respectively [21].
20 22 24 26 28 30 32 Weight gain in Afro-Caribbean people may
Body mass index (kg/m2) also be modified by the distribution of fat.
Namely, ectopic fat deposits in the liver and
Fig. 8.1 Prevalence of diabetes mellitus (%) in the
African Diaspora according to body mass index (Redrawn within the fascia surrounding the skeletal muscle
from data in Ref. [15]) (i.e. intermuscular adipose tissue or IMAT) may

WBISI Disposition index

** P <0.01
2.00 **



Marasmus Kwashiorkor Community Birth weight
survivors survivors controls controls

Fig. 8.2 Ratios after age, sex and BMI adjustment of glu- is a severe acute malnutrition that occurs with moderate
cose metabolism (WBISI and disposition index) of wasting (<60 % weight-for-age) and nutritional oedema.
kwashiorkor survivors, community controls and birth (3) Marasmus is a severe acute malnutrition occurring
weight controls compared to those of marasmus survivors. with severe wasting (<60 % weight-for-age) and no
Notes: (1) ratios represent the relative differences in each oedema. (4) WBISI is the whole-body insulin sensitivity
outcome between the kwashiorkor group and controls index. (5) Disposition index is derived from oral glucose
with the marasmus (comparator) group. (2) Kwashiorkor tolerance testing Data are from Ref. [44]
8 Diabetes in the Caribbean 131

be independent risk factors for insulin resistance adjusting for body composition, a one-unit
and diabetes. Afro-Caribbean women have large increase in physical activity level (i.e. expending
depots of subcutaneous fat and better adiponectin 590670 kJ or about 20 min of brisk walking
levels than men [22], so the roles of total fat, cen- daily) was associated with a 20-fold reduction in
tral adiposity and fat topology remain to be delin- the risk of diabetes [28].
eated to see if they would explain some of the
excess risk of diabetes in Caribbean women. Role of Developmental Factors
There is a dearth of data about hepatic steatosis in It is well established that low birth weight is asso-
the region. African Americans have greater IMAT ciated with the development of NCDs in later life
compared to Caucasians even after matching for [31, 32]. Birth weight is a crude marker or sum-
the total body fat and skeletal muscle mass [23]. mation of intrauterine growth, which in turn is
Afro-Tobagonian men have high amounts of determined by genetic factors, maternal body
intramuscular fat infiltration as measured by composition, maternal nutrition and placental
peripheral quantitative CT scan. The degree of sufficiency. So, children with low birth weight
IMAT is positively correlated with glycaemia, are more likely to have experienced growth
and this may also be modified by a family history restraint due to intrauterine nutritional restriction
of diabetes. Much of the variability in IMAT may or much less commonly, a genetic predisposition
be genetic, as the residual heritability (due to to low birth weight [33]. However, the associa-
additive genetic effects) was 35 % [24]. IMAT tion of birth weight with type 2 diabetes is J or
may thus be of equal or greater importance than U shaped, i.e. the prevalence of diabetes is
central adiposity [25]. The ectopic deposition of increased in individuals at both extremes of birth
lipids may be higher in obese Afro-Caribbean weight. The mechanisms underlying this rela-
women than in obese white women since African- tionship are not clear. However, both beta-cell
American women have higher rate of fatty acid dysfunction [34, 35] and insulin resistance [34
uptake and higher expression of fatty acid- 36] in childhood and adulthood may occur at the
transporting proteins [26]. extremes of birth weight. Other pathophysiologi-
Physical activity seems to be declining while cal mechanisms involved in low-birth-weight
sedentarism is increasing as part of the epidemio- individuals include hypothalamic-pituitary-
logic transition. The levels of physical activity in adrenal axis activation, visceral adiposity, changes
Barbadian young people are comparable to in adipocytokines and altered appetite. Large-for-
American youth [27], and both are low. There is gestational-age children are more likely to be the
also sexual dimorphism in physical activity as offspring of glucose-intolerant mothers. Thus,
Caribbean women generally report lower levels of they experience intrauterine hyperglycaemia
activity. The importance of this is underscored in causing fuel-mediated teratogenesis, which per se
urban Jamaica where severe or energetic physical may induce insulin resistance and type 2 diabetes
activity was uncommon in men, but non-existent in later life. This was demonstrated in the interna-
in women [28]. The more recent ecological study, tional Hyperglycemia and Adverse Pregnancy
Modelling the Epidemiologic Transition (METS), Outcome (HAPO) study which included more
using accelerometry confirmed that moderate-to- than 1200 Afro-Barbadian mothers [37].
vigorous activity only occurred for 12 min daily Interestingly, fasting glucose was inversely
in Caribbean women and was lower than African related to birth weight in peri-pubertal Jamaican
populations, but similar to American women [29]. boys, but directly associated with girls [38]. This
While men had moderate-to-vigorous activity sexual dimorphism might be due to girls being
about 10 min more than the women [30], they intrinsically more insulin resistant as argued by
were still relatively inactive although they some [39]. However, girls also have an earlier
remained lean. The protective effect of physical onset of puberty (an insulin-resistant state).
activity for incident diabetes is similar to other Earlier menarche and greater breast development
populations in a cross-sectional analysis. So, after in Jamaican children were associated with higher
132 M.S. Boyne

fasting glucose even after adjusting for current incident glucose tolerance more than worsening
BMI or prior growth rates [40]. In fact, fasting insulin resistance [45].
glucose increased by 0.6 mmol/L for each year Other developmentally influenced mecha-
reduction in the age of menarche. nisms can also be in play. Hypothalamic-pituitary-
In Jamaican children, shortness at birth and adrenal axis activation by early life events could
increased current weight are independent predic- account for some of the association with glycae-
tors of insulin resistance, as measured by 2-h mia. Hence, lower birth size and earlier gesta-
insulin levels [41] and increased glycosylated tional age are associated with higher nocturnal
haemoglobin levels [42]. However, data from a cortisol, which in turn is associated with lower
longitudinal birth cohort showed no relationship glucose effectiveness in adulthood [46]. Chronic
of birth size with insulin resistance measured in inflammation of adipose tissue has also been
childhood [38] or in youth [43]. Probably, post- implicated. As an example, faster growth in the
natal growth plays a role in the development of first 6 months of life is associated with higher
insulin resistance. So, children with faster post- serum adiponectin levels in later life [47]. This
natal growth during childhood (i.e. from ages 2 to implies that growth faltering in early infancy may
8 years) had greater insulin resistance, as mea- lead to hypoadiponectinaemia in later life, which
sured by HOMA, in later life [38]. in turn is associated with insulin resistance and
Prenatal factors may play a role in beta-cell glucose intolerance. Finally, there may be an
dysfunction in later life as seen in animal models. interaction with socio-economic status, i.e. young
So, greater maternal weight gain in pregnancy adults whose mothers had lower socio-economic
and smaller birth size are associated with reduced status during pregnancy have more adverse out-
beta-cell function in youth [43]. This phenome- comes associated with low birth size compared to
non is more clearly seen if severe acute malnutri- those from higher status [48]. This may imply the
tion in infancy also occurs. Namely, if children action of other unknown environmental factors
with low birth weight are weaned early onto low- and possibly even endocrine disruptors.
calorie feeds, they are more likely to develop The implication of these observations is
marasmus (typified by severe wasting, i.e. important for Caribbean people. At present the
weight-for-age <60 % but without nutritional average birth weight is ~3.1 kg, but there are a
oedema). Adult survivors of marasmus have significant number of low-birth-weight and mac-
more impaired glucose intolerance (odds ratio rosomic babies, as well as obese mothers. If these
10.9) than age, sex and BMI-matched commu- children have growth faltering in early infancy
nity controls [44]. They have marked beta-cell (the first 6 months of life), or rapid growth in late
dysfunction as measured by an oral disposition infancy/childhood, they are at an increased risk
index, but were only marginally more insulin of type 2 diabetes. Also, when this second gen-
resistant. eration of women later conceive, their offspring
It is therefore possible that intrauterine growth may be exposed to a hyperglycaemic intrauterine
restriction and/or postnatal undernutrition may environment (and thus be born macrosomic) or
impair the development of beta cells during this experience placental insufficiency (and thus be
plastic period of islet development during infancy born small for gestational age). Both conditions
leading to epigenetic changes that decrease insu- would increase the risk of glucose intolerance in
lin and PDX-1 gene expression. This would result the third generation.
in reduced beta-cell mass and glucose-stimulated
insulin secretion in later life. Historically, up to Role of Other Metabolic Factors
5 % of Caribbean populations experienced child- Other metabolic factors may be involved in the
hood severe acute malnutrition up to the 1950s, development of type 2 diabetes, such as
and some of these individuals may contribute to adipocytokines and oxidative stress. In Caribbean
the burden of diabetes today. In Jamaican adults, people, hypoadiponectinaemia is associated with
declining beta-cell function is more of a driver of incident glucose intolerance (OR ~0.93) [49, 50].
8 Diabetes in the Caribbean 133

Heritability estimates of adiponectin suggest that common. At present, Caribbean clinics are
genetic factors also influence the interindividual swamped with classical type 2 cases.
variation in circulating adiponectin levels and Patients with AKPD have periods in which
therefore the risk of glucose intolerance [51]. they are insulin-requiring resulting in ketosis,
While oxidative stress (e.g. lipid peroxides especially during metabolic stresses, e.g. infec-
such as isoprostanes) [52] and inflammatory tions [57]. At other times, their need for insulin
markers (e.g. sialic acid and highly sensitive decreases such that reasonable glycaemic control
C-reactive protein) [53] may be involved in dia- can be obtained with only lifestyle modification
betic complications in Afro-Caribbean persons, and/or oral antidiabetic agents. Some persons
they may not be involved in the pathogenesis of have insulin resistance and others have experi-
type 2 diabetes [45], but these studies may be enced malnutrition in childhood [58]. It is con-
underpowered. Similarly, glutathione levels are ceivable that they may have reduced beta-cell
marginally lower in patients with type 2 diabetes, mass and glucose-stimulated insulin secretion
but are significantly low in diabetic persons with due to malnutrition in early life similar to the
microvascular complications [54]. marasmic child [44]. Their beta cells appear to be
Chronobiology may play a role in a sex- sensitive to catabolic conditions leading to tran-
specific manner. In Caribbean men, insufficient sient, decreased glucose-stimulated insulin secre-
sleep (i.e. <6 h) or excessive sleep (>10 h) was tion. The precise metabolic triggers (e.g.
associated with diabetes when adjusted for age, glucotoxicity, adipocytokines, non-esterified
BMI and family history of diabetes (OR of 2.7 fatty acids causing lipotoxicity) are not known
and 4.4, respectively). Surprisingly though, in [59].
women sleeping less than 6 h was associated with APKD may be a heterogeneous collection of
a reduced likelihood of diabetes (OR 0.4) [55]. different phenotypes with different degrees of
impaired beta-cell function and autoimmunity
(i.e. anti-GAD65 and anti-IA-2 antibodies) [57].
Atypical Ketosis-Prone Diabetes In an African-American series of AKPD which
(AKPD)/Type 1B Diabetes also contained Afro-Caribbean patients, about
half had no evidence of autoimmunity (A-) with
In 1955, Hugh-Jones published one of the first preserved beta-cell function (+), while 22 %
descriptions of diabetes in the Caribbean where were A--, 17 % were A + - and 11 % were
he observed that type 2 diabetes was more com- A + + [60]. We do not have comparative data in
mon than type 1 diabetes in Jamaica [56]. He also Caribbean persons, but in one study of diabetes in
described an unusual variant which he called Caribbean youth, who were not necessarily
J-type diabetes J standing for Jamaica. In selected on the basis of having ketosis-prone dia-
more recent times, there have been other acro- betes, 30 % were A-+, 41 % were A-- and 39 %
nyms for J- type such as atypical diabetes, phasic were A + - or A + + [59, 61]. As such, some per-
insulin-dependent diabetes, ketosis-prone diabe- sons who have less autoimmunity, i.e. antibody
tes, type 3 diabetes, ketosis-prone type 2 diabetes negative but with preserved beta-cell reserve,
and Flatbush diabetes. This variant was associ- demonstrate a clinical course more in keeping
ated with insulin resistance, phasic dependency with type 2 diabetes despite having periods of
of insulin and a lean phenotype. It most resem- ketosis [62]. Persons with positive autoantibodies
bles type 1B diabetes in the WHO classification. tend to eventually need insulin therapy, while
It is more common in non-white populations, and persons with preserved beta-cell function may
marginal nutritional status may play a role in have periods of insulin independence [57].
some persons. In the Caribbean, there seems to A few candidate genes have been examined to
be fewer cases of AKPD conceivably because explain this variant of diabetes, but no genome-
improved nutrition in the Caribbean over the past wide association studies have been done to date.
several decades is making undernutrition less A missense mutation Gly574Ser in the transcrip-
134 M.S. Boyne

tion factor HNF-1 was thought to be a marker of North Africa region. There is some variation in
AKPD in African-American children [63]. the rates even within countries, depending on the
However, this candidate mutation was not signifi- methods used to diagnose diabetes (fasting glu-
cant in Afro-Caribbean patients [64]. Other cose vs. oral glucose tolerance testing, 1985
investigators working with other ethnic groups WHO diagnostic criteria vs. the 1997 criteria).
found that variants in HNF-1 and HNF-4 are Table 8.1 also gives a breakdown by country
unlikely to be major contributors to the pathogen- where logistic regression models were used to
esis of type 1B diabetes [65], but this is contro- produce smoothed age-specific prevalence rates
versial [66]. for adults aged 2079 years, which can allow for
country-to-country comparisons. Many countries
Maturity-Onset Diabetes of Youth have not carried out national surveys, and their
(MODY) prevalence rates are extrapolated from other
Like most of the world, MODY is rarely seen in Caribbean countries with a similar demographic
the Caribbean. Even though families demonstrat- profile. Clearly it is important for each country to
ing multigenerational inheritance of diabetes and have their data on the burden of disease.
other characteristics consistent with early-onset Naturally, incidence rates of diabetes are also
type 2 diabetes have been identified, evidence of high, although there is some difference based on
autoimmunity or sequence variants in MODY-1 ethnicity. In Jamaica, which is predominantly of
to MODY-6 genes were absent [67]. Insulin pro- African ancestry, the rates are 15 and 20 per 100
moter factor-1 (IPF-1) mutations in familial person-years for men and women, respectively
early-onset diabetes mellitus in Trinidadians [21]. However in Trinidad, where approximately
have been described [68]. 35 % of the population are of Indian ancestry and
another 35 % are of African, the incidence of dia-
betes is higher in Indo-Trinidadians. The rates of
Diagnosis of Diabetes Indo-Trinidadian men were 24 per 1000 person-
and Prediabetes in the Region years compared to Afro-Trinidadian men (13 per
1000 person-years) [17]. This was also similar for
Type 2 Diabetes Indo-Trinidadian women (23 per 1000 person-
years) and Afro-Trinidadian women (14 per 1000
In the 1950s and 1960s, diabetes was uncommon person-years) [17]. On account of the high inci-
in the Caribbean with rates <3 % [6971]. For dence rates, the IDF has been predicted that the
example, in 1958, the prevalence was 1.4 % after prevalence of diabetes in the Caribbean will
screening 2325 adults for glycosuria in Trinidad increase by 3760 % over the period by 2035. The
[72] and 0.73 % among 958 Jamaican adults [71]. epidemic is set to continue as there is a significant
Notably, women with diabetes outnumber men as reservoir of prediabetes. Impaired fasting glucose
far back as 1962 in Trinidad [69], even though occurs in ~3 % [74] and impaired glucose toler-
surprisingly they were not more overweight than ance in 13.7 % [10]. Of note, most surveys did not
the men. perform oral glucose tolerance testing, so most of
By the mid-1970s, PAHO sounded the alarm the regions estimate by the IDF (13.2 %) is actu-
for a looming epidemic of diabetes and NCDs in ally extrapolated from one late-1990s Jamaican
the region [73]. Since then, the prevalence of dia- survey [10]. Remarkably, there are little rural-
betes has increased significantly, and the urban differences [74] although there are differ-
Caribbean, like many developing countries, has ences by income status (lower income having
an epidemic of obesity and diabetes. The higher prevalence). Undiagnosed diabetes is a
International Diabetes Federation (IDF) esti- heavy burden for the region and will increase the
mates that the age-adjusted prevalence in 2013 morbidity, mortality and loss of human capital.
was 9.6 % for the Caribbean which is the second About 25 % of people with diabetes in the
highest in the world after the Middle East and Caribbean are unaware of their status [74, 75].
8 Diabetes in the Caribbean 135

Youth with Type 2 Diabetes Gestational Diabetes Mellitus (GDM)

Although cases of type 2 diabetes have been The epidemiologic transition would be expected
reported in youth, prevalence data is missing in to increase the risk for gestational diabetes, as
many countries. Anecdotally, these cases have well as the number of women who enter preg-
been increasing with the increasing girth of the nancy with type 2 diabetes. About 413 % of
regions youth. In the US Virgin Islands, the inci- Jamaican women of child-bearing age have dia-
dence of type 2 diabetes in youth age 19 years betes [10, 74]. There has been an apparent
rose significantly between 2001 (5.3/100,000) and increase in the incidence of GDM in Trinidad.
2010 (12.5/100,000) giving an age-adjusted annual Preliminary data show that from 2005 to 2007,
incidence rates of 9.6 per 100,000 [76]. In the prevalence of GDM increased from 1.7 to
Trinidadian schoolchildren, urine screening had a 6.7 % with a mean prevalence of 4.3 % [82].
positive predictive value of 65 % for detecting dia- Unknown genetic factors may play a role. A fam-
betes, and it showed a prevalence of 104/100,000 ily history of early-onset autosomal-dominant
while 75/100,000 had IGT [77]. This would, of type 2 diabetes seems to increase the risk of
course, be an underestimate compared to blood GDM in Jamaican women (12 % vs. 1.5 % in con-
screening. In a small Jamaican study, type 2 diabe- trols, i.e. OR 9.0) which would have implications
tes occurred in a third of cases of diabetes diag- for screening [83].
nosed in youth less than age 25 years [78]. Affected Women with prior GDM are at very high risk
youth were more likely to be female, older at diag- for incident type 2 diabetes. So, among
nosis, obese and have a higher blood pressure Trinidadian women with prior gestational diabe-
when compared to those with type 1 diabetes. tes, 62 % develop diabetes and 17 % had IGT
However, the strongest predictor of type 2 diabetes within 7 years [84]. These data suggest that
was obesity measured by BMI [78]. 1018 % of women convert to diabetes per year,
and the conversion to IGT occurs in another
35 %/year. If confirmed, these rates are extraor-
Type 1 Diabetes dinarily high compared to other regions of the
world [84].
Like many regions, type 1 diabetes is uncommon
in the Caribbean. Precise prevalence and inci-
dence data are sparse in many countries, but data Genomic Landscape of Diabetes
exist for Bahamas and the Eastern Caribbean. in the Region (Local Data, if
The prevalence was very high (31/100,000) in Available, or Best Approximation)
Bahamian youth <age 24 years with an incidence
rate of 10.1/100,000 in those age 014 years [79]. With its unique history of migration and coloni-
Incident rates are 45 per 100,000 person-years sation, it is no surprise that the Caribbean is an
in persons with African ancestry in the Eastern admixed region. Most persons claim African
Caribbean [80, 81] rising to a peak in the US ancestry, and one study using 28 ancestry
Virgin Islands (15.3 per 100,000 person-years) informative markers (AIMs) found that 8490 %
[76, 81] which may reflect their higher degree of had West African ancestry, 1012 % European
genetic admixture. The rates show some seasonal ancestry and 03 % Native American ancestry
variation and a tendency for a secular increase, [1]. Another study using 416 AIMs found Afro-
which augurs for unknown environmental factors Barbadians were ~77 % African, 16 % European
being part of the aetiology such as the accelerator and 7 % Asian [85]. However, Tobagonians had
hypothesis [80]. About half of patients with type the lowest rate of admixture (<6 %) [86].
1 diabetes are type 1A (with anti-GAD65 and/or Spanish Caribbean populations may have more
IA-2 antibodies positivity), and half are type 1B Native American ancestry [87], and this may play
(i.e. autoantibody negative) [61]. a role in the development of incident type 1 diabe-
136 M.S. Boyne

tes in these countries. Genetic susceptibility to Candidate-gene approaches for insulin resis-
type 1 diabetes is determined by a combination of tance have not been very successful. The PC-1
HLA-DQ and DRB1 genes (or a gene in linkage (ENPP1) K121Q polymorphism is not signifi-
disequilibrium with it). In a Cuban sample, a one- cantly associated (Colin A. McKenzie, personal
unit change in European admixture proportion communication, 2015). Variants in peroxisome
was associated with a 5.7-fold risk for type 1 dia- proliferator-activated receptor gamma (PPARG),
betes [88]. The HLA alleles DQA1*0501, *0301 and the obesity-associated gene, FTO, have not
DQB1*0201 and DRB1*0301, *0401 were sus- been well investigated in Caribbean populations.
ceptibility alleles, while DRB1*1501, The Trp64Arg mutation of the beta3-adrenergic
DQA1*0102/3 and DQB1*0602 were protective. receptor has been associated with hyperglycae-
In Afro-Jamaican patients, DRB1*03-DQ2/ mia and obesity in women, but not in men [94].
DRB1*04-DQ8, DRB1*0401-DQ8 and Persons of African ancestry have more fat
DRB1*0408-DQ8 genotypes increase the risk of infiltration of skeletal muscle fat than Europeans.
type 1 diabetes. The DRB1*1503-DQ6 and Since this is a heritable polygenic trait, a
DRB1*03-DQA1*0401-DQB1*0402 haplotypes candidate-gene study examined non-synonymous
were protective alleles. This pattern in the coding variants in carnitine palmitoyltransferase-
Jamaicans is different from the protective and pre- 1B (CPT1B). CPT1B is an enzyme that regulates
disposing haplotypes on European populations skeletal muscle mitochondrial beta oxidation of
[89]. Clearly more work is needed in this area. long-chain fatty acids. The G531L and I66V vari-
For type 2 diabetes, there have been a few ants were associated with ectopic fat infiltration in
studies utilising a candidate-gene approach as the skeletal muscle among 1774 older men from
well as genome-wide association studies, the population-based Tobago Health Study [95].
although some argue that genetic factors play Genome-wide association studies by the
only a minor role among Caribbean populations Genetic Investigation of ANthropometric Trait
[90]. A family history of diabetes in any first- (GIANT) consortium showed several single
degree relative (parent, sibling) or in a grandpar- nucleotide polymorphisms associated with adi-
ent is associated with a two- to fourfold increased posity. Using the phenotype, waist/hip ratio after
risk of diabetes [10, 91]. A family history of dia- adjusting for body mass index, they identified
betes is probably a summary statement for sev- several SNPs (49 loci of which 33 were new), and
eral interactive related genetic, but there is limited the effect size was stronger in women among 20
data on specific genetic factors involved in per- of these SNPs. These SNPs localised to pathways
sons of Caribbean origin. involved in adipogenesis, angiogenesis, tran-
Among the candidate genes, the transcription scriptional regulation, white adipose tissue dif-
factor, TCF7L2, which is involved in beta-cell ferentiation, insulin resistance, adipose
dysfunction, was associated with glucose intoler- inflammation (including adiponectin) and fat
ance in 385 Afro-Caribbean persons living in the topography in the 14,371 individuals of
UK [92]. This seems persistent among nonmigrant non-European ancestry, which also included
populations as among ~1000 Jamaicans, TCF7L2 2437 Afro-Jamaican individuals [96, 97].
was associated with decreased beta-cell function
(as measured by HOMA-%B) (unpublished data).
Mutations in the ATP-sensitive potassium Complications of Diabetes
channel (KCNJ11) in the beta cell are also asso- in the Region
ciated with impaired glucose-stimulated insulin
secretion. The common variant E23K was sig- Health Disparities in Complications
nificantly associated with type 2 diabetes in Indo-
Trinidadians (OR = 1.8) along with a few other People of African ancestry including Afro-
(novel) missense mutations A94P and R369C and Caribbean populations have lower rates of myo-
S118L (in an Afro-Trinidadian) [93]. cardial infarction compared to Caucasians [98,
8 Diabetes in the Caribbean 137

99], but conversely stroke [100] and renal failure nificant health disparities. In Barbados, diabetes
[101] rates are higher. It is not clear why such accounted for an excess mortality of 42 %, and
disparities exist, but genetic factors [98], coexist- there was a 9 % increase in all-cause mortality for
ing cardiometabolic risk factors [100], access to each 1 % increase in A1c [91]. However, there is
care and intensity of therapeutic control of risk dearth in data in other Caribbean countries about
factors are potential culprits. There is limited specific macrovascular complications and their
data on genetic factors influencing complications mortality. Cardiovascular disease is present in
in diabetic Caribbean people. Persons of African almost 60 % of hospitalised diabetic Jamaicans
ancestry have higher systolic blood pressures, but and is more frequent among women [106]. Silent
lower triglyceride and total cholesterol levels MI may occur in a quarter of Guadeloupian
[99]. Hypertension occurs in 52 % and 35 % of patients especially if they have left ventricular
diabetic Jamaican women and men, respectively hypertrophy [107]. The atrial natriuretic peptide
[10], and there may be less nocturnal dipping of rs5065 (2,238T>C) C allele seems to exert a pro-
blood pressure [102]. Tobacco use is relatively tective effect (24 % vs. 41 %, OR 0.5) in a rela-
low in diabetic persons [10] compared to devel- tively small study in Guadeloupe [108].
oped countries. As a result, these co-morbid car- Diabetes-related mortality is not limited to
diovascular differences may explain some of the older age groups, as 38 % of deaths occurred in
higher rates of stroke and renal failure (which are people under the age of 60 [109, 110]. Mortality
sensitive to blood pressure), and the less athero- in people with type 1 diabetes in the US Virgin
genic lipid profile may reduce the risk of coro- Islands is high, as cumulative survival was 98 %
nary artery disease. Notably, youth with diabetes at 10 years, but fell to 73 % at 20 years. This high
have a more adverse cardiometabolic pattern and incidence rate of T1D in the US Virgin Islands
thus may be more at risk of complications [103], may be partially responsible for the high mortal-
as was also seen in the SEARCH study. ity rate also seen [111].

Mortality and Macroangiopathy Diabetic Foot Disease

(Coronary Heart Disease and Stroke) and Amputations

The leading cause of death in the Caribbean is Diabetic foot disease has major public health
ischemic heart disease, and stroke is the second of consequences for the region. Approximately one
which diabetes is a major cause [104]. Also, dia- of every eight patients in diabetes clinics had a
betes is the leading causes of disability-adjusted major foot complication (amputation, ulcers,
life years. Mortality rates from diabetes have infection). Factors associated with these compli-
increased dramatically, i.e. 63 % from 1980 to cations were neuropathy (OR 9.3), high blood
1990. Diabetes-attributable macrovascular com- pressure (OR 7.9) and longer duration of diabetes
plications are affected by ethnicity. The (OR 1.32) [112]. Diabetic foot disease accounted
population-attributable mortality of Indo- for 30 % of admissions in Barbados and 89 % of
Trinidadians is 2.96.9 times higher than other diabetes-related admissions [113]. In fact,
ethnic groups [98, 105], and most of this is due to Trinidad uses 0.4 % of their gross domestic prod-
diabetes-induced cardiovascular disease. The age- uct solely to treat patients hospitalised for dia-
adjusted death rates due to diabetes in 2000 were betic foot infections [114].
25, 46, 56, 58 and 108 per 100,000 world standard Amputations for diabetic feet are disturbingly
population for Suriname, Bahamas, Barbados, high and are among the highest in the world.
Jamaica and Trinidad, respectively, according to Barbados has the most detailed data where the
PAHO (Health Statistics for the Americas, 2006 1-year incidence of lower extremity amputations
Edition, PAHO). In North America, the rates are was 936 per 100,000 populations (557 per
less than 16 per 100,000, so these data show sig- 100,000 for minor amputations and 379 per
138 M.S. Boyne

100,000 for major amputations) [115]. Women antihypertensive agents lowered the risk by a
had higher amputation rates than those reported half [123]. The risk of retinopathy increased by
for American Indians, and independent risk fac- 30 % for each 1 % increase of A1c [123].
tors were poor footwear (2.7-fold increased risk), Maculopathy was seen in 48 % of Jamaicans
elevated HbA1c (each increase of 1 % led to a with type 2 diabetes 30 years ago [124] and
40 % increase in amputations), peripheral neu- probably remains as high. In a high-risk clinic,
ropathy and peripheral vascular disease [115]. the frequency of diabetic retinopathy was 78 %;
Postamputation mortality rates at 1 and 5 years 30 % had background retinopathy, and 50 % of
are 31 % and 56 %, respectively, which are mostly the eyes had proliferative retinopathy of which
due to sepsis and cardiovascular disease the 34 % had tractional retinal detachments [125].
highest reported worldwide [116]. Besides retinopathy, the burden of glaucoma and
Most of these disparities in the Caribbean are cataracts is also high [91]. These data clearly
probably due to social influences (e.g. inappro- demonstrate the need for retinopathy screening
priate footwear) and healthcare delivery. glycaemic control and blood pressure control in
However, there may be biological factors. For prevention.
example, a persistent inflammatory response in
diabetic ulcers is mostly due to overproduction of
TNF [117]. This persistent inflammation can be Nephropathy
aggravated by low levels of the TNF-receptor as
is seen in persons with the allele, TRAPS In a Jamaican diabetes clinic, about 22 % had
P46L. Interestingly, the allele frequency of eGFR <60 ml/min/1.73 m2, 21 % had moderate
TRAPS P46L in the Barbadian population was albuminuria and 62 % had severe albuminuria
9.5 %, which is 30 times higher than Caucasian [126]. Data from the Caribbean Renal Registry
populations, and could contribute to these high showed that diabetes accounted for ~28 % of the
amputation rates [118]. It is also possible that cases of end-stage renal disease [127]. In many
haemorheological factors, such as increased countries, diabetic persons of African ancestry
plasma viscosity and fibrinogen, may also play a have higher rates of nephropathy. One possible
role in the development of the diabetic foot [119]. explanation is that the increased oxidative stress
seen in Afro-Caribbean persons may be involved
in increased renal damage [52]. Accordingly,
Retinopathy Caribbean patients with microangiopathy have
glutathione deficiency that is probably due to
Some of the best data are from the Barbados Eye reduced synthesis and increased irreversible utili-
Study which showed that the prevalence of reti- sation by non-glycaemic mechanisms [54].
nopathy was approximately 29 % [120]. Most Nephropathy may also be more common in
persons had mild disease while ~8 % had moder- atypical ketosis-prone diabetes [128] for unclear
ate changes, and 1 % had severe retinopathy. reasons.
Clinically significant macular oedema was found
in 9 % of those with diabetes. In the same cohort
after 4 years, the incidence was 32 % in those Hyperglycaemic Crises
with known diabetes at baseline and 21 % in per-
sons with newly diagnosed diabetes [121]. About 60 % of persons admitted with diabetic
Clinically significant macular oedema developed ketoacidosis have type 2 diabetes [129]. The prev-
in 5 %. This increased to 40 % after 9 years and alence of hyperglycaemic crises in the general dia-
the incidence of macular oedema was ~9 % betic population is unknown. Mortality was 7 %
[122]. Increased systolic blood pressure was a for ketoacidosis, 20 % for hyperosmolar hypergly-
risk factor, and over 9 years the relative risk was caemia and 25 % for persons with the mixed keto-
1.3 for every 10 mmHg increase, while using acidosis/hyperosmolar syndrome [129].
8 Diabetes in the Caribbean 139

Other Complications that the The health of the region is the wealth of
the region in their 2001 Nassau Declaration.
Depression was present in 18 % of type 2 diabetic Subsequently, CARICOM reaffirmed this com-
patients, and older age, Indian ancestry, women mitment in the 2007 Declaration of Port of Spain
and co-morbidities were risk factors [130]. There to begin public health initiatives focused against
are little or no data on other possible complica- NCDs including diabetes. These initiatives would
tions of diabetes such as erectile dysfunction, include primary prevention as well as providing
osteopenia, vascular dementia, dental disease and adequate secondary care. They specifically
reduced lung function in the Caribbean. included antitobacco legislation, promoting
healthy diets that also include indigenous foods
for children, and promote nationwide increases in
Coordination and Delivery physical activity and national commissions to
of Diabetes Care Services coordinate NCD interventions. CARICOM lob-
bied the United Nations to have a high-level
Several studies have shown that cardiometabolic meeting on NCDs which finally occurred in
control of Caribbean diabetic patients is subopti- 1920 September 2011 [138]. This call for a
mal [74] in both private and public sectors [131], coordinated macroeconomic, multisectoral
although this may be improving [132, 133]. In approach is a credit and desperately needed,
fact, less than half attain an A1c goal of <7 %, although there was no significant funding alloca-
and only a small fraction achieve the trifecta of tion by CARICOM, which limited the implemen-
target A1c, blood pressure and lipids [134]. tation of secondary care. Each nation had to find
Patient factors and healthcare system inadequa- the resources to achieve the goals. So, more cul-
cies probably contributed to these issues. turally sensitive incentives to promote healthy
Unfortunately many primary care clinicians are balanced nutrition, opportunities to incorporate
not following any guidelines [135], and, if so, leisure-time physical activities as well as increas-
periodic repetition and reinforcement [136] are ing activity during routine daily living are still
needed to maintain competency. needed. Many countries adopted socialised,
Therefore it is no surprise that the loss of government-subsidised approaches by providing
human capital and the economic costs to the diabetic medications and glucose monitoring
region are staggering. By one conservative esti- supplies. Although many Caribbean countries are
mate, diabetes costs the English-speaking coun- small, access to care is still an issue in inner
tries in 2000 US$218.1 million in direct costs, cities, and rural areas are mountainous and many
US$812.4 million in indirect costs and US$687 times relatively inaccessible. Some countries
million per capita in direct costs along with 5555 have considered mobile clinics to improve access.
diabetes-related deaths [137]. Diabetes and hyper- There have been some efforts to standardise the
tension cost Bahamas 9.1 % of its GDP, Barbados clinical care of persons with diabetes. The
>5 %, Jamaica 6 % and Trinidad and Tobago 8 %. Caribbean Public Health Agency (CARPHA) is
These data are old and are probably a very conser- the new single regional public health agency for the
vative underestimate especially with the ageing of Caribbean, established in 2011 by CARICOM and
the Caribbean and its growing obesity problem. began operation in January 2013. Its predecessor
This growing economic burden is borne by indi- organisation, the Caribbean Health Research
viduals and the governments. Many individuals Council (CHRC), in partnership with PAHO pub-
do not have the required financial resources as lished clinical guidelines for managing diabetes in
many have no retirement savings, pension, lack the Caribbean primary care setting in 2006. The
health insurance, and unemployment is high in metabolic target goals were similar to the IDF
several middle- and lower-income countries. guidelines although there was a more stringent
As a consequence of this threat to Caribbean LDL-cholesterol goal of <1.8 mmol/L. The degree
health and development, CARICOM proposed of uptake and implementation appeared to be inad-
140 M.S. Boyne

equate by the regions clinicians, and proper multi- Table 8.2 Glycaemic indices of some staples commonly
eaten in the Caribbean according to their method of cook-
disciplinary diabetes management teams were not
well established in many territories, leading to gaps
in holistic care. Newly updated evidence-based Food Boiled Other Preparation
guidelines by CARPHA are to be released in 2016 Cassava 94 (baked)
Breadfruit 47 72 (roasted)
that will try to address areas such as diabetes edu-
Green banana 37 35 (fried)
cation, the central role of diabetes self-management
Ripe plantain 66 90 (fried)
and the prevention of diabetic complications. The
Green plantain 39 40 (fried)
therapeutic approach will be similar to the ADA/
Dasheen 72
EASD guidelines. CARPHA, along with regional
White yam 75 80 (roasted)
universities, is attempting to establish a structured
Lucea yam 74 77 (roasted)
surveillance system for screening and monitoring
Yellow yam 68 80 (roasted)
of NCDs. Systematic screening for risk factors and
Negro yam 73 73 (roasted)
strategies aimed at reducing risk factors for obesity
Sweet potato 46 76 (fried), 82 (roasted)
and type 2 diabetes among schoolchildren are Irish potato 59 70 (fried), 83 (baked)
urgently needed [77, 139]. Eddoes 61
Pumpkin 66
Roti 65 (baked)
Non-pharmacological Management Data adapted from [142, 143]

clinical trial showed that incorporating low-
As mentioned, the nutritional transition towards a intermediate GI indigenous foods into patients
more Westernised diet has helped to drive the meal planning lowered A1c by 0.84 % and also
epidemic. Several surveys have shown energy- improved chronic inflammation (hsCRP), homo-
dense, high-fat, salt are common in the region. cysteine, HDL-C and triglyceride levels [144].
Unfortunately, the Caribbean suffers from a lack
of trained nutritionists and dietetic professionals.
In response to this need, PAHO and the Caribbean Diabetes Education
Food and Nutrition Institute produced a manual
on the nutritional management of obesity, diabe- Diabetes knowledge tends to be poor which
tes and hypertension although it is now outdated, affects their care [145147] but this seems to be
especially with the introduction of so many need slowly improving. Certified diabetes educators
foodstuffs to the market over the decade. are needed although there have been attempts to
Encouragingly, Caribbean patients are inter- fill this void over the past decade by the non-
ested in non-pharmacological management since profit organisation, Diabetes Educators of the
culturally it is deemed more natural. However, Caribbean. Lay persons in communities trained
food security in certain regions poses a threat to in diabetes prevention and care can provide an
this. There is great interest in using locally avail- alternative to traditional point of care of access
able foods like sweet potatoes and yams, but the for diabetes management [148]. Using lay diabe-
glycaemic indices of these foods were not well tes facilitators in Jamaica was associated with an
documented until more recently (Table 8.2) A1c reduction of 0.5 % [149, 150].
[140143]. The mode of cooking can also affect
the glycaemic index (GI). Frying and roasting are
more popular although these methods increase Exercise
the GI compared to boiling. Promoting boiling
can therefore be an important method for control- Physical activity is low especially among women,
ling postprandial spikes. In support, a randomised but there has been increasing interest in recreational
8 Diabetes in the Caribbean 141

physical activity with more gyms, personal train- Metformin has been used in the region for more
ers, yoga, Pilates and use of public spaces and than 40 years and is the most prescribed agent.
parks for walking. A randomised trial showed Sulphonylureas are commonly used because of
6 months of hatha yoga exercise or conventional their low cost although anecdotally there is a sig-
physical therapy improved fasting blood glucose, nificant amount of symptomatic hypoglycaemia
lipid profile and oxidative stress markers and anti- among the elderly. Most oral agents and insulins
oxidant status in Jamaican patients [151]. are available in the region, and access is determined
by the formulary of the countries. Many clinicians
follow the ADA/EASD guidelines, which are
Nontraditional/Complementary incorporated in the updated CARPHA guidelines,
Techniques and thus use medications (metformin, sulphonyl-
ureas, DPP-4 inhibitors, pioglitazone, acarbose,
There is great interest by patients to use tradi- SGLT2 inhibitors, GLP-1 receptor agonists, basal
tional herbs for glycaemic control [152]. Annatto insulin, NPH insulin, regular insulin, rapid-acting
(Bixa orellana) [153], bitter yam [154], cashews insulin analogues) depending on patient factors
(Anacardium occidentale) [155] and chilli pep- (cost, risk of hypoglycaemia, co-morbid disease,
per (Capsicum frutescens which contains capsa- side effects). In government formularies, metfor-
icin) [156] have mild hypoglycaemic activity. min, sulphonylureas, acarbose, NPH insulin and
Many use guinea hen weed (Petiveria alliacea), regular insulin are used most often.
but it appears to have little hypoglycaemic effect
[157]. This area of natural products remains of
interest for local research. Translating Primary Prevention
of Type 2 Diabetes

Bariatric Surgery Stemming the tide of diabetes requires primary

prevention. Of course, this is the most cost-
Bariatric surgery is an emerging treatment in the effective strategy, and the rise of NCDs threatens
Caribbean but only a very small number of per- the fragile economies of these island nations.
sons have used it. Of these, 85 % had resolution CARPHA instituted the annual Caribbean
of their diabetes, and 15 % were able to reduce Wellness Day (CWD) in 2008, in an effort to
their medication with improvements in quality of strengthen public, private and civil society partner-
life and few complications [158]. ships and to promote multi-country, multisectoral
activities in support of wellness [159]. Culturally
sensitive initiatives are needed but there are few
Tobacco Cessation studies investigating the relationships between the
formation of NCD policy and culture [160].
Tobacco use is relatively low in the Caribbean
compared to other regions, especially among
women. More recently, many countries have Women
passed antitobacco legislation as part of the
CARICOMs drive to reduce the impact of NCDs. Women need special attention as most studies in
the Caribbean identify a higher burden of disease in
women as well as poorer glycaemic control. In the
Rational Selection of Antidiabetes many other regions, women are at a similar or
Medications lower risk of type 2 diabetes than men, even when
obesity is higher in women. This female excess
The CARPHA and national guidelines emphasise may be due to a much greater excess of obesity and
the use of metformin as initial pharmacotherapy. other risk factors in women (e.g. physical activity).
142 M.S. Boyne

However, there may be other factors involved

which need further research. These findings have Policy-Development). The mandate of its
major implications for preventive policies since Research, Training and Policy Development Unit
gender-sensitive messages are needed [18]. is to promote research for health, advise govern-
ments and other stakeholders on research for
health, strengthen national and regional health
Children and Youth research systems, develop mechanisms to support
priority research and promote the sharing of the
The burden of cardiometabolic risk starts early and regions scientific output. Some of its activities
is seen in children. In a recent study [161], about include the administration of research grants, the
one third of Jamaican adolescents were over- hosting of the annual scientific meetings, the deliv-
weight, the girls were less physically active, over ery of training workshops along with the produc-
80 % had 3 risk factors for type 2 diabetes and tion of accompanying manuals (basic and
cardiovascular disease and one third had the meta- advanced research skills, research ethics, grant
bolic syndrome. Thus, interventions are needed to writing and monitoring and evaluation) and the
educate children to reduce their risk. A simple promotion of essential national health research.
measure for risk stratification is the waist circum- The research agenda is packed as it also includes
ference. A waist circumference of 82 cm in young HIV/AIDS and emerging infectious diseases, but
men had better sensitivity to identify insulin resis- nutrition and NCDs including diabetes are priori-
tance than the IDF standard of 94 cm (45 % vs. ties. However, there is no specific set of research
14 %), but the IDF standard of 80 cm for young priorities for diabetes itself. The regions universi-
women remains a reasonable threshold [162]. ties have accelerated their research output for dia-
betes although each investigator acts independently
in determining their research agenda. Investigations
Early Life Interventions into natural products, epidemiological surveys,
behavioural studies, developmental origins and
If cardiometabolic risk can start during foetal complications are more common areas. There is a
life, it raises the issue of optimising the health of paucity of clinical trials for the unique needs of the
women of child-bearing age. Most specifically, region and this needs expanding.
this would involve ensuring a near-normal BMI
prior to entering pregnancy. Interventions that
start in pregnancy involving physical activity and Future Directions: Unmet Needs,
nutrition may actually be too late in the patho- Unanswered Questions,
physiological chain, and thus the effect sizes are Unquestioned Answers
likely to be small, if any. However, prevention of
excessive weight gain in pregnancy, similar to the The preceding description of the state of diabetes
Institute of Medicine guidelines, may be useful. in the Caribbean would highlight that there are
Universal screening for gestational diabetes is several areas that need attention in the future.
not present in all countries, and this should also Some are listed below in no particular order of
be instituted. significance:

Greater efficiencies and synergism of research

Organising and Conduction efforts across the region
Diabetes Research in the Region Greater understanding of the role of obesity
and fat distribution
The Caribbean Public Health Agency is also More understanding of the nature of atypical
responsible for conducting relevant research on ketosis-prone diabetes and how to effectively
public health priorities in the Caribbean (see http:// intervene
8 Diabetes in the Caribbean 143

Expanding foot care and better vascular type 1 diabetes and islet cell autoantibody-positive
subjects in a population with a low incidence of type
1 diabetes. Autoimmunity. 2007;40:5405.
Expanding knowledge about the role of early 7. Bartholomew C, Cleghorn F. Retroviruses in the
life factors and possible primary prevention Caribbean. Bull Pan Am Health Organ. 1989;23:
methods appropriate for the region 7680.
8. Smikle MF, Wright-Pascoe R, Barton EN, Dowe G,
Mechanistic understanding of ethnomedicine
Gilbert DT, Choo-Kang E, et al. Autoantibodies,
and relevant clinical studies in promising nat- human T lymphotropic virus type 1 and type 1 dia-
ural products betes mellitus in Jamaicans. West Indian Med
Better systems to collect and track national J. 2002;51:1536.
9. Costa GC, Azevedo R, Gadelha SR, Kashima SH,
data about the NCDs and their complications
Muricy G, Olavarria VN, et al. Polymorphisms at
Greater development of diabetes team GLUT1 gene are not associated with the development
approaches as well as ensuring training of pro- of TSP/HAM in Brazilian HTLV-1 infected individ-
fessionals who can be retained in the region as uals and the discovery of a new polymorphism at
GLUT1 gene. J Med Virol. 2009;81:5527.
there is significant brain drain by migration to
10. Wilks R, Rotimi C, Bennett F, McFarlane-Anderson
the North N, Kaufman JS, Anderson SG, et al. Diabetes in the
Useful models of health reform and funding Caribbean: results of a population survey from Spanish
Devising mechanisms for improved efficien- Town, Jamaica. Diabet Med. 1999;16:87583.
11. Luke A, Cooper RS, Prewitt TE, Adeyemo AA,
cies of health delivery
Forrester TE. Nutritional consequences of the
Gender-sensitive approaches for tailoring African diaspora. Annu Rev Nutr. 2001;21:4771.
public health interventions 12. Luke A, Durazo-Arvizu R, Rotimi C, Prewitt TE,
What social re-engineering is necessary for Forrester T, Wilks R, et al. Relation between body
mass index and body fat in black population samples
primary prevention
from Nigeria, Jamaica, and the United States. Am
To understand the interactions between cul- J Epidemiol. 1997;145:6208.
ture and health policy formation 13. Cooper R, Rotimi C, Ataman S, McGee D,
Health economists are needed to help frame Osotimehin B, Kadiri S, et al. The prevalence of
hypertension in seven populations of west African
some of the regions data into useful advice
origin. Am J Public Health. 1997;87:1608.
for policymakers 14. Durazo-Arvizu RA, Luke A, Cooper RS, Cao G,
Dugas L, Adeyemo A, et al. Rapid increases in obe-
sity in Jamaica, compared to Nigeria and the United
States. BMC Public Health. 2008;8:133.
References 15. Cooper RS, Rotimi CN, Kaufman JS, Owoaje EE,
Fraser H, Forrester T, et al. Prevalence of NIDDM
1. Benn-Torres J, Bonilla C, Robbins CM, Waterman among populations of the African diaspora. Diabetes
L, Moses TY, Hernandez W, et al. Admixture and Care. 1997;20:3438.
population stratification in African Caribbean popu- 16. Franco M, Bilal U, Ordunez P, Benet M, Morejon A,
lations. Ann Hum Genet. 2008;72:908. Caballero B, et al. Population-wide weight loss and
2. Global, regional, and national age-sex specific all- regain in relation to diabetes burden and cardiovas-
cause and cause-specific mortality for 240 causes of cular mortality in Cuba 19802010: repeated cross
death, 19902013: a systematic analysis for the Global sectional surveys and ecological comparison of sec-
Burden of Disease Study 2013. Lancet. 385:11771. ular trends. BMJ. 2013;346:f1515.
3. Wilks R, Bennett F, Forrester T, McFarlane- 17. Miller GJ, Maude GH, Beckles GL. Incidence of
Anderson N. Chronic diseases: the new epidemic. hypertension and non-insulin dependent diabetes
West Indian Med J. 1998;47 Suppl 4:404. mellitus and associated risk factors in a rapidly
4. Sargeant LA, Wilks RJ, Forrester TE. Chronic dis- developing Caribbean community: the St James sur-
eases facing a public health challenge. West Indian vey, Trinidad. J Epidemiol Community Health.
Med J. 2001;50 Suppl 4:2731. 1996;50:497504.
5. Sarmiento L, Cubas-Duenas I, Cabrera-Rode 18. Sobers-Grannum N, Murphy MM, Nielsen A, Guell
E. Evidence of association between type 1 diabetes C, Samuels TA, Bishop L, et al. Female gender is a
and exposure to enterovirus in Cuban children and social determinant of diabetes in the Caribbean: a
adolescents. MEDICC Rev. 2013;15:2932. systematic review and meta-analysis. PLoS One.
6. Sarmiento L, Cabrera-Rode E, Lekuleni L, Cuba I, 2015;10:e0126799.
Molina G, Fonseca M, et al. Occurrence of enterovi- 19. Sargeant LA, Boyne MS, Bennett FI, Forrester TE,
rus RNA in serum of children with newly diagnosed Cooper RS, Wilks RJ. Impaired glucose regulation
144 M.S. Boyne

in adults in Jamaica: who should have the oral glu- uptake in men aged 70 years in relation to size at
cose tolerance test? Rev Panam Salud Publica. birth. Diabetologia. 1998;41:11338.
2004;16:3542. 32. McCance DR, Pettitt DJ, Hanson RL, Jacobsson LT,
20. El Mabchour A, Delisle H, Vilgrain C, Larco P, Knowler WC, Bennett PH. Birth weight and non-
Sodjinou R, Batal M. Specific cut-off points for insulin dependent diabetes: thrifty genotype, thrifty
waist circumference and waist-to-height ratio as pre- phenotype, or surviving small baby genotype? BMJ.
dictors of cardiometabolic risk in Black subjects: a 1994;308:9425.
cross-sectional study in Benin and Haiti. Diabetes 33. Gluckman PD, Hanson MA, Cooper C, Thornburg
Metab Syndr Obes Targets Ther. 2015;8:51323. KL. Effect of in utero and early-life conditions on
21. Sargeant LA, Bennett FI, Forrester TE, Cooper RS, adult health and disease. N Engl J Med. 2008;359:
Wilks RJ. Predicting incident diabetes in Jamaica: the 6173.
role of anthropometry. Obes Res. 2002;10:7928. 34. Li C, Johnson MS, Goran MI. Effects of low birth
22. Boyne MS, Bennett NR, Cooper RS, Royal-Thomas weight on insulin resistance syndrome in caucasian
TY, Bennett FI, Luke A, et al. Sex-differences in adi- and African-American children. Diabetes Care.
ponectin levels and body fat distribution: longitudi- 2001;24:203542.
nal observations in Afro-Jamaicans. Diabetes Res 35. Soto N, Bazaes RA, Pena V, Salazar T, Avila A,
Clin Pract. 2010;90:e336. Iniguez G, et al. Insulin sensitivity and secretion are
23. Albu JB, Kovera AJ, Allen L, Wainwright M, Berk related to catch-up growth in small-for-gestational-
E, Raja-Khan N, et al. Independent association of age infants at age 1 year: results from a prospective
insulin resistance with larger amounts of intermus- cohort. J Clin Endocrinol Metab. 2003;88:364550.
cular adipose tissue and a greater acute insulin 36. Crowther NJ, Trusler J, Cameron N, Toman M, Gray
response to glucose in African American than in IP. Relation between weight gain and beta-cell secre-
white nondiabetic women. Am J Clin Nutr. 2005; tory activity and non-esterified fatty acid production
82:12107. in 7-year-old African children: results from the Birth
24. Miljkovic-Gacic I, Wang X, Kammerer CM, Gordon to Ten study. Diabetologia. 2000;43:97885.
CL, Bunker CH, Kuller LH, et al. Fat infiltration in 37. Metzger BE, Lowe LP, Dyer AR, Trimble ER,
muscle: new evidence for familial clustering and Chaovarindr U, Coustan DR, et al. Hyperglycemia
associations with diabetes. Obesity (Silver Spring). and adverse pregnancy outcomes. N Engl J Med.
2008;16:185460. 2008;358:19912002.
25. Miljkovic-Gacic I, Gordon CL, Goodpaster BH, 38. Boyne MS, Osmond C, Fraser RA, Reid M, Taylor-
Bunker CH, Patrick AL, Kuller LH, et al. Adipose Bryan C, Soares-Wynter S, et al. Developmental ori-
tissue infiltration in skeletal muscle: age patterns and gins of cardiovascular risk in Jamaican children: the
association with diabetes among men of African Vulnerable Windows Cohort study. Br J Nutr. 2010;
ancestry. Am J Clin Nutr. 2008;87:15905. 104:102633.
26. Bower JF, Davis JM, Hao E, Barakat HA. Differences 39. Murphy MJ, Metcalf BS, Voss LD, Jeffery AN,
in transport of fatty acids and expression of fatty acid Kirkby J, Mallam KM, et al. Girls at five are intrinsi-
transporting proteins in adipose tissue of obese black cally more insulin resistant than boys: the
and white women. Am J Physiol Endocrinol Metab. Programming Hypotheses Revisited The EarlyBird
2006;290:E8791. Study (EarlyBird 6). Pediatrics. 2004;113:826.
27. Prochaska JJ, Sallis JF, Griffith B, Douglas 40. Boyne MS, Thame M, Osmond C, Fraser RA, Gabay
J. Physical activity levels of Barbadian youth and L, Taylor-Bryan C, et al. The effect of earlier puberty
comparison to a U.S. sample. Int J Behav Med. on cardiometabolic risk factors in Afro-Caribbean
2002;9:36072. children. J Pediatr Endocrinol Metab. 2014;27:
28. Boyne MS, Gaskin P, Luke A, Wilks RJ, Bennett FI, 45360.
Younger N, et al. Energetic determinants of glucose 41. Bennett F, Watson-Brown C, Thame M, Wilks R,
tolerance status in Jamaican adults. Eur J Clin Nutr. Osmond C, Hales N, et al. Shortness at birth is asso-
2004;58:16668. ciated with insulin resistance in pre-pubertal
29. Luke A, Bovet P, Plange-Rhule J, Forrester TE, Jamaican children. Eur J Clin Nutr. 2002;56:
Lambert EV, Schoeller DA, et al. A mixed ecologic- 50611.
cohort comparison of physical activity & weight 42. Forrester TE, Wilks RJ, Bennett FI, Simeon D,
among young adults from five populations of African Osmond C, Allen M, et al. Fetal growth and cardio-
origin. BMC Public Health. 2014;14:397. vascular risk factors in Jamaican schoolchildren.
30. Dugas LR, Bovet P, Forrester TE, Lambert EV, BMJ. 1996;312:15660.
Plange-Rhule J, Durazo-Arvizu RA, et al. 43. Chin-Harty LA, Royal-Thomas T, Thompson DS,
Comparisons of intensity-duration patterns of physi- Osmond C, Forrester TE, Boyne MS. Role of prena-
cal activity in the US, Jamaica and 3 African coun- tal factors and postnatal growth on insulin sensitivity
tries. BMC Public Health. 2014;14:882. and beta-cell function in Afro-Caribbean youth: vul-
31. McKeigue PM, Lithell HO, Leon DA. Glucose toler- nerable Windows Cohort Study. J Dev Orig Health
ance and resistance to insulin-stimulated glucose Dis. 2015;6:S12. (abstract).
8 Diabetes in the Caribbean 145

44. Francis-Emmanuel PM, Thompson DS, Barnett AT, 56. Hugh-Jones P. Diabetes in Jamaica. Lancet.
Osmond C, Byrne CD, Hanson MA, et al. Glucose 1955;269:8917.
metabolism in adult survivors of severe acute malnu- 57. Balasubramanyam A, Nalini R, Hampe CS,
trition. J Clin Endocrinol Metab. 2014;99:223340. Maldonado M. Syndromes of ketosis-prone diabetes
45. Boyne MS, Sargeant LA, Bennett FI, Wilks RJ, mellitus. Endocr Rev. 2008;29:292302.
Cooper RS, Forrester TE. Isoprostanes, a marker of 58. Morrison EY, Ragoobirsingh D. J type diabetes
lipid peroxidation, may not be involved in the devel- revisited. J Natl Med Assoc. 1992;84:6038.
opment of glucose intolerance. Diabetes Res Clin 59. Boyne MS. Diabetes in the Caribbean: trouble in
Pract. 2007;76:14951. paradise. Insulin. 2009;4:94105.
46. Thompson DS, Ferguson TS, Wilks RJ, Phillips DI, 60. Maldonado M, Hampe CS, Gaur LK, DAmico S,
Osmond C, Samms-Vaughan M, et al. Early-life fac- Iyer D, Hammerle LP, et al. Ketosis-prone diabetes:
tors are associated with nocturnal cortisol and glu- dissection of a heterogeneous syndrome using an
cose effectiveness in Afro-Caribbean young adults. immunogenetic and beta-cell functional
Clin Endocrinol (Oxf). 2015;82:3528. classification, prospective analysis, and clinical out-
47. Boyne MS, Thompson DS, Osmond C, Fraser RA, comes. J Clin Endocrinol Metab. 2003;88:50908.
Thame MM, Taylor-Bryan C, et al. The effect of 61. Tulloch-Reid MK, Boyne MS, Choo-Kang EG,
antenatal factors and postnatal growth on serum adi- Parkes R, Wright-Pascoe RA, Barton EN, et al.
ponectin levels in children. J Dev Orig Health Dis. Autoantibodies in Caribbean youth with diabetes
2013;4:31723. mellitus. Hum Antibodies. 2008;17(34):5762.
48. Ferguson TS, Younger-Coleman NO, Tulloch-Reid 62. Mauvais-Jarvis F, Sobngwi E, Porcher R, Riveline
MK, Knight-Madden JM, Bennett NR, Samms- JP, Kevorkian JP, Vaisse C, et al. Ketosis-prone type
Vaughan M, et al. Birth weight and maternal socio- 2 diabetes in patients of sub-Saharan African origin:
economic circumstances were inversely related to clinical pathophysiology and natural history of beta-
systolic blood pressure among Afro-Caribbean cell dysfunction and insulin resistance. Diabetes.
young adults. J Clin Epidemiol. 2015;68:10029. 2004;53:64553.
49. Ezenwaka CE, Kalloo R. Caribbean female patients 63. Boutin P, Gresh L, Cisse A, Hara M, Bell G, Babu S,
with type 2 diabetes mellitus have lower serum lev- et al. Missense mutation Gly574Ser in the transcrip-
els of adiponectin than nondiabetic subjects. Neth tion factor HNF-1alpha is a marker of atypical dia-
J Med. 2005;63:649. betes mellitus in African-American children.
50. Bennett NR, Boyne MS, Cooper RS, Royal-Thomas Diabetologia. 1999;42:3801.
TY, Bennett FI, Luke A, et al. Impact of adiponectin 64. Mauvais-Jarvis F, Boudou P, Sobngwi E, Riveline
and ghrelin on incident glucose intolerance and on JP, Kevorkian JP, Villette JM, et al. The polymor-
weight change. Clin Endocrinol (Oxf). 2009;70: phism Gly574Ser in the transcription factor HNF-
40814. 1alpha is not a marker of adult-onset ketosis-prone
51. Miljkovic-Gacic I, Wang X, Kammerer CM, Bunker atypical diabetes in Afro-Caribbean patients.
CH, Wheeler VW, Patrick AL, et al. Genetic deter- Diabetologia. 2003;46:7289.
mination of adiponectin and its relationship with 65. Aguilera E, Casamitjana R, Ercilla G, Oriola J,
body fat topography in multigenerational families of Nicoletti F, Gomis R, et al. Clinical characteristics,
African heritage. Metabolism. 2007;56:2348. beta-cell function, HLA class II and mutations in
52. Mehrotra S, Ling KL, Bekele Y, Gerbino E, Earle MODY genes in non-paediatric subjects with Type 1
KA. Lipid hydroperoxide and markers of renal dis- diabetes without pancreatic autoantibodies. Diabet
ease susceptibility in African-Caribbean and Med: J Br Diabet Assoc. 2005;22:13743.
Caucasian patients with Type 2 diabetes mellitus. 66. Stride A, Ellard S, Clark P, Shakespeare L, Salzmann
Diabet Med. 2001;18:10915. M, Shepherd M, et al. Beta-cell dysfunction, insulin
53. Nayak BS, Roberts L. Relationship between inflam- sensitivity, and glycosuria precede diabetes in hepa-
matory markers, metabolic and anthropometric vari- tocyte nuclear factor-1alpha mutation carriers.
ables in the Caribbean type 2 diabetic patients with Diabetes Care. 2005;28:17516.
and without microvascular complications. J Inflamm 67. Mills JL, Irving RR, Choo-Kang EG, Wright-Pascoe
(Lond). 2006;3:17. R, McLaughlin W, Mullings AA, et al.
54. Lutchmansingh FK, Bennett FI, Badaloo AV, Multigenerational inheritance and clinical character-
McFarlane-Anderson N, Gordon-Strachan GM, istics of three large pedigrees with early-onset type 2
Wright-Pascoe R, et al. Glutathione metabolism in diabetes in Jamaica. Rev Panam Salud Publ Pan Am
persons with type 2 diabetes. West Indian Med J Publ Health. 2010;27:43541.
J. 2012;61:24. (abstract). 68. Cockburn BN, Bermano G, Boodram LL,
55. Cumberbatch CG, Younger NO, Ferguson TS, Teelucksingh S, Tsuchiya T, Mahabir D, et al. Insulin
McFarlane SR, Francis DK, Wilks RJ, et al. Reported promoter factor-1 mutations and diabetes in
hours of sleep, diabetes prevalence and glucose control Trinidad: identification of a novel diabetes-
in Jamaican adults: analysis from the Jamaica lifestyle associated mutation (E224K) in an Indo-Trinidadian
survey 20072008. Int J Endocrinol. 2011;716214. family. J Clin Endocrinol Metab. 2004;89:9718.
146 M.S. Boyne

69. Pyke DA, Wattley GH. Diabetes in Trinidad. West with a family history of autosomal dominant type 2
Indian Med J. 1962;11:226. diabetes. Rev Panam Salud Publ Pan Am J Publ
70. Poon-King T, Henry MV, Rampersad F. Prevalence Health. 2008;23:8591.
and natural history of diabetes in Trinidad. Lancet. 84. Ali Z, Alexis SD. Occurrence of diabetes mellitus
1968;291:15560. after gestational diabetes mellitus in Trinidad.
71. Tulloch JA, Johnson HM. A pilot survey of the inci- Diabetes Care. 1990;13:5279.
dence of diabetes in Jamaica. West Indian Med 85. Murray T, Beaty TH, Mathias RA, Rafaels N, Grant
J. 1958;7:1346. AV, Faruque MU, et al. African and non-African
72. Wright HB, Taylor B. The incidence of diabetes in a admixture components in African Americans and an
sample of the adult population in South Trinidad. African Caribbean population. Genet Epidemiol.
West Indian Med J. 1958;7:12333. 2010;34:5618.
73. Litvak J. Diabetes mellitus: a challenge for the coun- 86. Miljkovic-Gacic I, Ferrell RE, Patrick AL,
tries of the region. Bull Pan Am Health Organ. Kammerer CM, Bunker CH. Estimates of African,
1975;9:31724. European and Native American ancestry in Afro-
74. Cunningham-Myrie C, Younger-Coleman N, Caribbean men on the island of Tobago. Hum Hered.
Tulloch-Reid M, McFarlane S, Francis D, Ferguson 2005;60:12933.
T, et al. Diabetes mellitus in Jamaica: sex differences 87. Moreno-Estrada A, Gravel S, Zakharia F, McCauley
in burden, risk factors, awareness, treatment and JL, Byrnes JK, Gignoux CR, et al. Reconstructing
control in a developing country. Tropical Med Int the population genetic history of the Caribbean.
Health: TM IH. 2013;18:136578. PLoS Genet. 2013;9:e1003925.
75. Ferguson TS, Francis DK, Tulloch-Reid MK, 88. Diaz-Horta O, Cintado A, Fernandez-De-Cossio
Younger NO, McFarlane SR, Wilks RJ. An update ME, Nazabal M, Ferrer A, Roca J, et al. Relationship
on the burden of cardiovascular disease risk factors of type 1 diabetes to ancestral proportions and HLA
in Jamaica: findings from the Jamaica Health and DR/DQ alleles in a sample of the admixed Cuban
Lifestyle Survey 20072008. West Indian Med population. Ann Hum Biol. 2010;37:77888.
J. 2011;60:4228. 89. Heward JM, Mijovic CH, Kelly MA, Morrison E,
76. Washington RE, Orchard TJ, Arena VC, Laporte Barnett AH. HLA-DQ and DRB1 polymorphism
RE, Tull ES. Incidence of type 1 and type 2 diabetes and susceptibility to type 1 diabetes in Jamaica. Eur
in youth in the U.S. Virgin Islands, 20012010. J Immunogenet. 2002;29:4752.
Pediatr Diabetes. 2013;14:2807. 90. Cruickshank JK, Mbanya JC, Wilks R, Balkau B,
77. Batson YA, Teelucksingh S, Maharaj R, Singh V, McFarlane-Anderson N, Forrester T. Sick genes,
Balkaran S, Cockburn B. Screening for diabetes in sick individuals or sick populations with chronic dis-
schoolchildren in Trinidad, West Indies. Paediatr Int ease? The emergence of diabetes and high blood
Child Health. 2013;33:3741. pressure in African-origin populations. Int
78. Tulloch-Reid MK, Boyne MS, Smikle MF, Choo- J Epidemiol. 2001;30:1117.
Kang EG, Parkes RH, Wright-Pascoe RA, et al. 91. Hennis A, Wu SY, Nemesure B, Li X, Leske
Clinical and laboratory features of youth onset type MC. Diabetes in a Caribbean population: epidemio-
2 diabetes in Jamaica. West Indian Med J. 2010; logical profile and implications. Int J Epidemiol.
59:1318. 2002;31:2349.
79. Peter SA, Johnson R, Taylor C, Hanna A, Roberts P, 92. Humphries SE, Gable D, Cooper JA, Ireland H,
McNeil P, et al. The incidence and prevalence of Stephens JW, Hurel SJ, et al. Common variants in
type-1 diabetes mellitus. J Natl Med Assoc. 2005;97: the TCF7L2 gene and predisposition to type 2 diabe-
2502. tes in UK European Whites, Indian Asians and Afro-
80. Jordan OW, Lipton RB, Stupnicka E, Cruickshank Caribbean men and women. J Mol Med.
JK, Fraser HS. Incidence of type I diabetes in people 2006;84:100514.
under 30 years of age in Barbados, West Indies, 93. Boodram LG, Miyake K, Hayes MG, Bell GI,
19821991. Diabetes Care. 1994;17:42831. Cockburn BN. Association of the KCNJ11 variant
81. Tull ES, Jordan OW, Simon L, Laws M, Smith DO, E23K with type 2 diabetes in Indo-Trinidadians.
Vanterpool H, et al. Incidence of childhood-onset West Indian Med J. 2011;60:6047.
IDDM in black African-heritage populations in the 94. McFarlane-Anderson N, Bennett F, Wilks R, Howell
Caribbean. The Caribbean African Heritage IDDM S, Newsome C, Cruickshank K, et al. The Trp64Arg
Study (CAHIS) Group. Diabetes Care. 1997;20: mutation of the beta3-adrenergic receptor is associ-
30910. ated with hyperglycemia and current body mass
82. Clapperton M, Jarvis J, Mungrue K. Is gestational index in Jamaican women. Metabolism.
diabetes mellitus an important contributor to meta- 1998;47:61721.
bolic disorders in Trinidad and tobago? Obstet 95. Miljkovic I, Yerges LM, Li H, Gordon CL,
Gynecol Int. 2009;2009:289329. Goodpaster BH, Kuller LH, et al. Association of
83. Irving RR, Mills JL, Choo-Kang EG, Morrison EY, the CPT1B gene with skeletal muscle fat infiltra-
Kulkarni S, Wright-Pascoe R, et al. The burden of tion in Afro-Caribbean men. Obesity. 2009;17:
gestational diabetes mellitus in Jamaican women 1396401.
8 Diabetes in the Caribbean 147

96. Shungin D, Winkler TW, Croteau-Chonka DC, Caribbean: an update. Diabetes Res Clin Pract.
Ferreira T, Locke AE, Magi R, et al. New genetic 2014;103:22330.
loci link adipose and insulin biology to body fat dis- 110. Andrade F. Estimating diabetes and diabetes-free
tribution. Nature. 2015;518:18796. life expectancy in Mexico and seven major cities in
97. Locke AE, Kahali B, Berndt SI, Justice AE, Pers Latin America and the Caribbean. Rev Panam Salud
TH, Day FR, et al. Genetic studies of body mass Publ Pan Am J Publ Health. 2009;26:916.
index yield new insights for obesity biology. Nature. 111. Washington RE, Orchard TJ, Arena VC, LaPorte
2015;518:197206. RE, Secrest AM, Tull ES. All-cause mortality in a
98. Beckles GL, Miller GJ, Kirkwood BR, Alexis SD, population-based type 1 diabetes cohort in the
Carson DC, Byam NT. High total and cardiovascular U.S. Virgin Islands. D Diabetes Res Clin Pract.
disease mortality in adults of Indian descent in 2014;103:5049.
Trinidad, unexplained by major coronary risk fac- 112. Ferguson TS, Tulloch-Reid MK, Younger NO,
tors. Lancet. 1986;1:1298301. Wright-Pascoe RA, Boyne MS, McFarlane SR, et al.
99. McKeigue PM, Shah B, Marmot MG. Relation of Diabetic foot complications among patients attend-
central obesity and insulin resistance with high dia- ing a specialist diabetes clinic in Jamaica: preva-
betes prevalence and cardiovascular risk in South lence and associated factors. West Indian Med
Asians. Lancet. 1991;337:3826. J. 2013;62:21623.
100. Chaturvedi N. Ethnic differences in cardiovascular 113. Taylor Jr CG, Krimholtz M, Belgrave KC,
disease. Heart. 2003;89:6816. Hambleton I, George CN, Rayman G. The extensive
101. Earle KK, Porter KA, Ostberg J, Yudkin JS. Variation inpatient burden of diabetes and diabetes-related
in the progression of diabetic nephropathy according foot disease in Barbados. Clin Med. 2014;14:
to racial origin. Nephrol Dial Transplant. 2001;16: 36770.
28690. 114. Cawich SO, Islam S, Hariharan S, Harnarayan P,
102. Chaturvedi N, McKeigue PM, Marmot MG. Resting Budhooram S, Ramsewak S, et al. The economic
and ambulatory blood pressure differences in Afro- impact of hospitalization for diabetic foot infections
Caribbeans and Europeans. Hypertension. 1993;22: in a Caribbean nation. Permanente J. 2014;18:
906. e1014.
103. Tulloch-Reid MK, Boyne MS, Smikle MF, Choo- 115. Hennis AJ, Fraser HS, Jonnalagadda R, Fuller J,
Kang EG, Parkes RH, Wright-Pascoe RA, et al. Chaturvedi N. Explanations for the high risk of
Cardiovascular risk profile in Caribbean youth with diabetes-related amputation in a Caribbean popula-
diabetes mellitus. West Indian Med J. 2009;58: tion of black African descent and potential for pre-
21926. vention. Diabetes Care. 2004;27:263641.
104. Mortality GBD, Causes of Death C. Global, regional, 116. Hambleton IR, Jonnalagadda R, Davis CR, Fraser
and national age-sex specific all-cause and cause- HS, Chaturvedi N, Hennis AJ. All-cause mortality
specific mortality for 240 causes of death, 1990 after diabetes-related amputation in Barbados: a pro-
2013: a systematic analysis for the Global Burden of spective case-control study. Diabetes Care. 2009;32:
Disease Study 2013. Lancet. 2015;385:11771. 3067.
105. Miller GJ, Kirkwood BR, Beckles GL, Alexis SD, 117. Landis RC, Evans BJ, Chaturvedi N, Haskard
Carson DC, Byam NT. Adult male all-cause, cardio- DO. Persistence of TNFalpha in diabetic wounds.
vascular and cerebrovascular mortality in relation to Diabetologia. 2010;53:15378.
ethnic group, systolic blood pressure and blood glu- 118. Quimby KR, Greenidge AR, Hennis AJ, Harrison
cose concentration in Trinidad. West Indies Int DK, Landis RC. Tumor necrosis factor receptor-
J Epidemiol. 1988;17:629. associated periodic syndrome P46L and bilateral
106. Ferguson TS, Tulloch-Reid MK, Younger NO, amputation in diabetes. Rheumatology. 2010;
Boyne MS, Wright-Pascoe RA, Elliott VE, et al. 49:24545.
Cardiovascular disease among diabetic in-patients at 119. Vigilance JE, Reid HL. Glycaemic control influ-
a tertiary hospital in Jamaica. Diab Vasc Dis Res. ences peripheral blood flow and haemorheological
2010;7:2412. variables in patients with diabetes mellitus. Clin
107. Blanchet Deverly A, Amara M, Larifla L, Hemorheol Microcirc. 2005;33:33746.
Velayoudom-Cephise FL, Roques F, Kangambega P, 120. Leske MC, Wu SY, Hyman L, Li X, Hennis A,
et al. Silent myocardial ischaemia and risk factors in Connell AM, et al. Diabetic retinopathy in a black
a diabetic Afro-Caribbean population. Diabetes population: the Barbados Eye Study. Ophthalmology.
Metab. 2011;37:5339. 1999;106:18939.
108. Larifla L, Maimaitiming S, Velayoudom-Cephise 121. Leske MC, Wu SY, Hennis A, Nemesure B, Hyman
FL, Ferdinand S, Blanchet-Deverly A, Benabdallah L, Schachat A. Incidence of diabetic retinopathy in
S, et al. Association of 2238T>C polymorphism of the Barbados Eye Studies. Ophthalmology. 2003;
the atrial natriuretic peptide gene with coronary 110:9417.
artery disease in Afro-Caribbeans with type 2 diabe- 122. Leske MC, Wu SY, Hennis A, Nemesure B, Schachat
tes. Am J Hypertens. 2012;25:5247. AP, Hyman L, et al. Nine-year incidence of diabetic
109. Yisahak SF, Beagley J, Hambleton IR, Narayan KM, retinopathy in the Barbados Eye Studies. Arch
Atlas IDFD. Diabetes in North America and the Ophthalmol. 2006;124:2505.
148 M.S. Boyne

123. Leske MC, Wu SY, Hennis A, Hyman L, Nemesure 139. Batson YA, Teelucksingh S, Maharaj RG,
B, Yang L, et al. Hyperglycemia, blood pressure, and Cockburn BN. A cross-sectional study to deter-
the 9-year incidence of diabetic retinopathy: the mine the prevalence of obesity and other risk fac-
Barbados Eye Studies. Ophthalmology. 2005;112: tors for type 2 diabetes among school children in
799805. Trinidad, West Indies. Paediatr Int Child Health.
124. Moriarty BJ, Dunn DT, Moriarty AP. Diabetic macu- 2014;34:17883.
lopathy in a Jamaican population. Int Ophthalmol. 140. Bahado-Singh PS, Riley CK, Wheatley AO, Lowe
1989;13:3013. HI. Relationship between processing method and the
125. Mowatt L. Diabetic retinopathy and its risk factors at glycemic indices of ten sweet potato (Ipomoea bata-
the university hospital in Jamaica. Middle East Afr tas) cultivars commonly consumed in Jamaica.
J Ophthalmol. 2013;20:3216. J Nutr Metab. 2011;2011:584832.
126. Ferguson TS, Tulloch-Reid MK, Younger-Coleman 141. Riley CK, Bahado-Singh PS, Wheatley AO, Ahmad
NO, Wright-Pascoe RA, Boyne MS, Soyibo AK, MH, Asemota HN. Relationship between the physi-
et al. Prevalence of chronic kidney disease among cochemical properties of starches and the glycemic
patients attending a specialist diabetes clinic in indices of some Jamaican yams (Dioscorea spp.).
Jamaica. West Indian Med J. 2015;64:201. Mol Nutr Food Res. 2008;52:13726.
127. Soyibo AK, Barton EN. Report from the Caribbean 142. Bahado-Singh PS, Wheatley AO, Ahmad MH,
renal registry, 2006. West Indian Med J. 2007;56: Morrison EY, Asemota HN. Food processing meth-
35563. ods influence the glycaemic indices of some com-
128. Ragoobirsingh D, Bennett F, Morrison EY. Kidney monly eaten West Indian carbohydrate-rich foods.
function in phasic insulin dependent diabetes melli- Br J Nutr. 2006;96:47681.
tus in Jamaica. West Indian Med J. 1997;46:224. 143. Ramdath DD, Isaacs RL, Teelucksingh S, Wolever
129. Chung ST, Perue GG, Johnson A, Younger N, Hoo TM. Glycaemic index of selected staples commonly
CS, Pascoe RW, et al. Predictors of hyperglycaemic eaten in the Caribbean and the effects of boiling v.
crises and their associated mortality in Jamaica. crushing. Br J Nutr. 2004;91:9717.
Diabetes Res Clin Pract. 2006;73:18490. 144. Bahado-Singh PS, Wheatley AO, Osagie AU, Boyne
130. Frederick FT, Maharajh HD. Prevalence of depres- MS, Morrison EYS, Ahmad MH, et al. Low to
sion in type 2 diabetic patients in Trinidad and intermediate-glycemic-index Caribbean foods
Tobago. West Indian Med J. 2013;62:62831. reduce glycemia, cardiovascular and inflammatory
131. Wilks RJ, Sargeant LA, Gulliford MC, Reid ME, markers in overweight persons with type 2 diabetes.
Forrester TE. Management of diabetes mellitus in Curr Res Nutr Food Sci. 2015;3:3645.
three settings in Jamaica. Rev Panam Salud Publica. 145. Wint YB, Duff EM, McFarlane-Anderson N, O'Connor
2001;9:6572. A, Bailey EY, Wright-Pascoe RA. Knowledge, moti-
132. Mahabir D, Gulliford MC. Changing patterns of pri- vation and barriers to diabetes control in adults in
mary care for diabetes in Trinidad and Tobago over Jamaica. West Indian Med J. 2006;55:3303.
10 years. Diabet Med. 2005;22:61924. 146. Babwah F, Baksh S, Blake L, Cupid-Thuesday J,
133. Apparico N, Clerk N, Henry G, Seale J, Sealy R, Hosein I, Sookhai A, et al. The role of gender in
Ward S, et al. How well controlled are our type 2 compliance and attendance at an outpatient clinic for
diabetic patients in 2002? An observational study in type 2 diabetes mellitus in Trinidad. Rev Panam
North and Central Trinidad. Diabetes Res Clin Pract. Salud Publ Pan Am J Publ Health. 2006;19:7984.
2007;75:3015. 147. Adams OP, Carter AO. Knowledge, attitudes, prac-
134. Forde H, Marshall M, Cohall D, Nielsen tices, and barriers reported by patients receiving dia-
AL. Assessing predictors for sustainable management betes and hypertension primary health care in
of type 2 diabetes using evidence-based guidelines in Barbados: a focus group study. BMC Fam Pract.
public primary care in a predominantly Afro- 2011;12:135.
Caribbean population. Ethn Dis. 2014;24:46974. 148. Morrison E, Haye I, Amarakoon K, Whitbourne F,
135. Adams OP, Carter AO. Are primary care practitio- Kirlew J, Less L, et al. An interim report of an inter-
ners in Barbados following diabetes guidelines? a vention strategy for diabetes care in Jamaica. West
chart audit with comparison between public and pri- Indian Med J. 2001;50 Suppl 1:559.
vate care sectors. BMC Res Notes. 2011;4:199. 149. Less LA, Ragoorbirsingh D, Morrison EY, Boyne
136. Adams OP, Carter AO. Diabetes and hypertension M, Johnson P. Assessment of a community based
guidelines and the primary health care practitioner in Intervention for diabetes control in adults with type
Barbados: knowledge, attitudes, practices and barri- 2 diabetes mellitus in three parishes in Jamaica: par-
ers a focus group study. BMC Fam Pract. 2010;11:96. ticipant were assigned to interventions using random
137. Barcelo A, Aedo C, Rajpathak S, Robles S. The cost allocations. West Indian Med J. 2008;57 Suppl 1:39.
of diabetes in Latin America and the Caribbean. Bull 150. Less LA, Ragoobirsingh D, Morrison EY, Boyne M,
World Health Organ. 2003;81:1927. Johnson PA. A preliminary report on an assessment
138. Hassell T, Hennis A. The road to the United Nations of a community-based intervention for diabetes con-
High Level Meeting on chronic non-communicable trol in adults with type 2 diabetes. Fam Pract. 2010;27
diseases. West Indian Med J. 2011;60:3846. Suppl 1:i4652.
8 Diabetes in the Caribbean 149

151. Gordon LA, Morrison EY, McGrowder DA, Young 157. Christie SL, Levy A. Evaluation of the hypoglycae-
R, Fraser YT, Zamora EM, et al. Effect of exercise mic activity of Petiveria alliacea (Guinea Hen Weed)
therapy on lipid profile and oxidative stress indica- extracts in normoglycaemic and diabetic rat models.
tors in patients with type 2 diabetes. BMC West Indian Med J. 2013;62:68591.
Complement Altern Med. 2008;8:21. 158. Dan D, Harnanan D, Singh Y, Hariharan S,
152. Picking D, Younger N, Mitchell S, Delgoda Naraynsingh V, Teelucksingh S. Effects of bariatric
R. The prevalence of herbal medicine home use surgery on type-2 diabetes mellitus in a Caribbean
and concomitant use with pharmaceutical medi- setting. Int J Surg. 2011;9:38691.
cines in Jamaica. J Ethnopharmacol. 159. Samuels TA, Fraser H. Caribbean Wellness Day:
2011;137:30511. mobilizing a region for chronic noncommunicable
153. Russell KR, Omoruyi FO, Pascoe KO, Morrison disease prevention and control. Rev Panam Salud
EY. Hypoglycaemic activity of Bixa orellana extract Publ Pan Am J Publ Health. 2010;28:4729.
in the dog. Methods Find Exp Clin Pharmacol. 160. Samuels TA, Guell C, Legetic B, Unwin N. Policy
2008;30:3015. initiatives, culture and the prevention and control of
154. McAnuff MA, Harding WW, Omoruyi FO, Jacobs chronic non-communicable diseases (NCDs) in the
H, Morrison EY, Asemota HN. Hypoglycemic Caribbean. Ethn Health. 2012;17:63149.
effects of steroidal sapogenins isolated from 161. Barrett SC, Huffman FG, Johnson P, Campa A,
Jamaican bitter yam, Dioscorea polygonoides. Food Magnus M, Ragoobirsingh D. A cross-sectional
Chem Toxicol: Int J Published Br Ind Biol Res study of Jamaican adolescents risk for type 2
Assoc. 2005;43:166772. diabetes and cardiovascular diseases. BMJ Open.
155. Alexander-Lindo RL, Morrison EY, Nair 2013;3:e002817.
MG. Hypoglycaemic effect of stigmast-4-en-3-one 162. Tulloch-Reid MK, Ferguson TS, Younger NO, Van
and its corresponding alcohol from the bark of den Broeck J, Boyne MS, Knight-Madden JM, et al.
Anacardium occidentale (cashew). Phytother Res: Appropriate waist circumference cut points for iden-
PTR. 2004;18:4037. tifying insulin resistance in black youth: a cross sec-
156. Tolan I, Ragoobirsingh D, Morrison EY. Isolation tional analysis of the 1986 Jamaica birth cohort.
and purification of the hypoglycaemic principle Diabetol Metab Syndr. 2010;2:68.
present in Capsicum frutescens. Phytother Res:
PTR. 2004;18:956.
Diabetes in Indigenous Australians
and Other Underserved 9
Communities in Australia

Stephen Colagiuri

Introduction Services Scheme (NDSS) [2], the Australian

Health Survey (National Health Measures
Australias population has just reached 24 million Survey, 20112013) [3], and the AusDiab study
which is relatively small for its large land mass. (19992000) [4]. The Australian Health Survey
The population is diverse with over one in four found that using HbA1c (glycated hemoglo-
being born overseas and includes a unique bin 6.5 %) as the measure, 5.4 % of the popu-
Indigenous population which has lived here for lation over the age of 18 had diabetes, which
thousands of years. It is well recognized that dia- translates to almost 1 million people [3]. Using
betes and other chronic diseases are overrepre- fasting plasma glucose 7.0 mmol/l as the mea-
sented in these diverse groups. This review sure, 5.1 % of Australians aged 18 years and over
examines aspects of diabetes prevalence and care had diabetes. The majority were people known
in three population groups in Australia in which to have diabetes with approximately one newly
health outcomes are of particular concern and are diagnosed case of diabetes for every four already
a significant source of disadvantage: diagnosed cases. Diabetes was more common in
men than women (6.3 % compared with 3.9 %)
Australias Indigenous population of for both known diabetes (4.9 % compared with
Aboriginal and Torres Strait Islander people 3.4 %) and newly diagnosed diabetes (1.4 % com-
People from culturally and linguistically pared with 0.4 %) [3].
diverse backgrounds The 19992000 AusDiab (Australian Diabetes
People living in rural and remote areas and Obesity Lifestyle) study used an oral glucose
tolerance test (OGTT) to estimate that 7.5 % of
Australians over the age of 25 had diabetes,
Diabetes in Australia which equates to approximately 1.2 million peo-
ple [4]. A similar estimate comes from the NDSS,
It is estimated that more than 1.2 million a government-funded national scheme which
Australians are living with diabetes based on data provides registrants with subsidized consumables
from three sources [1]: the National Diabetes for self-monitoring blood glucose, insulin nee-
dles, and insulin pump consumables. The NDSS
has registered 1.2 million people with diabetes
S. Colagiuri, MD
and registers over 250 people every day [2].
Boden Institute of Obesity, Nutrition and Exercise,
University of Sydney, Sydney, NSW, Australia However, it is likely that all three sources
e-mail: underestimate the total number of people affected

Springer International Publishing Switzerland 2017 151

S. Dagogo-Jack (ed.), Diabetes Mellitus in Developing Countries and Underserved Communities,
DOI 10.1007/978-3-319-41559-8_9
152 S. Colagiuri

by diabetes. The Australian Health Survey did not Survey reported prediabetes in 4.0 % of adults
perform an OGTT. The AusDiab study was con- diagnosed on the basis of a fasting plasma glu-
ducted over 15 years ago, and during this time, cose of 6.16.9 mmol/l and 6.0 % based on an
there has been a considerable increase in the pro- HbA1c of 6.06.4 %. These data highlight the
portion of people who are overweight and obese. underestimation of prediabetes when using tests
Both these studies excluded children. Registration which do not specifically identify impaired glu-
with the NDSS is voluntary, and it is likely that a cose tolerance (IGT) [3].
number of people with diabetes are not registered. The greatest risk factor for diabetes is obesity.
While the contribution of many chronic dis- In 20112012, 63 % of Australians aged 18 years
eases to the national burden of disability is and over were overweight or obese, comprised of
decreasing, the diabetes burden continues to 35 % overweight and 28 % obese. The prevalence
increase and is predicted to become the largest of overweight and obesity has increased in
contributor to disability-adjusted life years by Australia over time, from 56 % in 1995 to 61 % in
2017 [5]. In 2011, diabetes was the underlying 20072008. Overweight and obesity vary with
cause of 3 % of all deaths and an underlying or age, with 74.9 % of people aged 6574 years
associated cause of 10 % of all deaths in Australia being overweight or obese compared with 36.4 %
[6]. There are a significant number of diabetes- of people aged 1824 years. More men were
related complications, including heart attacks, overweight or obese than women (70 % vs. 56 %).
strokes, limb amputations, blindness, kidney fail- However, looking at only those people who were
ure, nerve damage, and depression. It is estimated obese, rates are the same for men and women
that annually there are 840,000 hospitalizations (both 28 %). The proportion of people who are
for diabetes, 3500 people having dialysis due to obese has increased across all age groups over
diabetes, and 3500 lower-limb amputations due time, up from 19 % in 1995 to 28 % in 2011
to diabetes [7]. 2012. Between 1995 and 20112012, average
Diabetes impacts the individual, their family, weight increased 3.9 kg in men and 4.1 kg in
and society in general. People with diagnosed women. Rates of overweight and obesity in chil-
diabetes (approximately 5 % of Australians) dren and adolescents aged 517 have remained
account for 12 % of total health care costs in stable at 25 % [10].
Australia. For type 1 diabetes, total direct costs
amount to a minimum of A$570 million
per annum: A$4669 annually for a person with no Australias Indigenous People
complications, rising to A$16,698 per annum
once complications develop [8]. Type 2 diabetes The term Indigenous Australians refers to the
costs at least A$14.6 billion annually, 42 % of Aboriginal and Torres Strait Islander people of
which is attributed to direct medical costs. These Australia, descended from groups that existed
costs are projected to increase to A$30 billion by in Australia and surrounding islands prior to
2025 [9]. The cost pattern is similar in people European colonization in the late eighteenth cen-
with type 2 diabetes with the annual cost increas- tury. The time of arrival of the first Indigenous
ing from A$3500 for people without diabetes Australians is at least 40,000 years ago and may
complications to A$9600 for people with compli- even date back as far as 125,000 years. There is
cations [9]. Australia is fortunate in having the great diversity among different Indigenous com-
NDSS to offset some of these costs. munities and societies in Australia, each with its
In addition to diabetes, intermediate hypergly- own mixture of cultures, customs, and languages.
cemia (prediabetes) is also common. The The Indigenous population of Aboriginal and
AusDiab study found that 16.4 % of adults over Torres Strait Islanders make up approximately
the age of 25 (approximately 2.5 million people) 3 % of the entire population. The age profile
had prediabetes based on a standard 2 h OGTT of Australias Indigenous and non-Indigenous
[4]. More recent data from the Australian Health populations is considerably different with a
9 Diabetes in Indigenous Australians and Other Underserved Communities in Australia 153

larger proportion of young people and a smaller has higher rates of diabetes, significant prema-
proportion of older people (4 % vs. 15 % aged 65 ture mortality, and high rates of complications,
and over) among Indigenous Australians [11]. especially cardiovascular and renal disease.
Life expectancy at birth for Indigenous In addition, the diabetes burden is increased in
Australians in the period 20102012 was populations living in rural and remote areas of
69.1 years for males and 73.7 years for females Australia [14].
compared with 79.7 years for males and 83.1 years The 20122013 National Aboriginal and
for females for non-Indigenous Australians a Torres Strait Islander Health Measures Survey is
gap of 10.6 years for males and 9.5 years for the largest biomedical survey ever conducted in
females [12]. Between 20052007 and 2010 Indigenous Australians and included some 3300
2012, life expectancy at birth for Indigenous individuals aged 18 years and over across
males increased from 67.5 to 69.1 years and from Australia. In addition to collecting data on self-
73.1 to 73.7 years for Indigenous females [13]. reported diabetes, it included measurement of
In 20092013, the age-standardized mortality fasting plasma glucose and glycated hemoglobin.
rate for Indigenous Australians was 985 per The survey showed that 11.1 % of Indigenous
100,000, 1.7 times higher than the 585 per adults had diabetes 9.6 % with previously diag-
100,000 for non-Indigenous Australians. The rate nosed diabetes and 1.5 % with newly diagnosed
ratio was highest for the 3544 years age group diabetes, indicating approximately one newly
where the Indigenous mortality rate was 4.2 diagnosed case for every six diagnosed cases.
times that of the non-Indigenous rate [14]. After taking age differences into account,
Although there have been some improve- Indigenous people were more than three times as
ments, the overall health status of Indigenous likely as non-Indigenous people to have diabe-
Australians remains a concern. Smoking rates tes 3.6 times more likely to have known diabe-
have declined from 51 % to 44 % between 2002 tes and twice as likely to have newly diagnosed
and 20122013 for Indigenous Australians aged diabetes. Indigenous women were significantly
15 and over but remain high with a current 25 more likely than men to have diabetes [15].
percentage point gap between Indigenous and Diabetes prevalence among Indigenous peo-
non-Indigenous Australians. Smoking during ple increased with age. Rates were especially
pregnancy remains high at 50 %. Low birth high among those aged 55 years and over, with
weight rate for babies born to Indigenous moth- around one in every three people in this age group
ers is twice that of non-Indigenous mothers (11 % having diabetes (34.5 %), compared with 12 %
compared with 5 %). It is estimated that 51 % of among non-Indigenous Australians, a gap of 22
low birth weight births to Indigenous mothers is percentage points [15] (Fig. 9.1).
attributable to smoking, compared with 19 % for Although this overall age pattern was similar
non-Indigenous Australian mothers. If smoking to non-Indigenous Australians, diabetes tended to
rates for Indigenous pregnant women were the occur earlier. The prevalence of diabetes for
same as for other Australian mothers, it is esti- Indigenous people aged 3544 years was 9.0 %
mated that the proportion of low birth weight which is similar to that for non-Indigenous peo-
babies could be reduced by 26 % [14]. ple aged 5564 years (8.2 %). Similarly, the prev-
alence for those aged 4554 years was 17.8 %,
similar to that for those aged 6574 in the non-
Diabetes in Indigenous Australians Indigenous population (15.0 %). This pattern was
apparent for both known diabetes and newly
Prevalence diagnosed diabetes [15].
Minges et al. performed a systematic review
The burden of diabetes is not shared equally in of diabetes prevalence in Indigenous Australians
Australia, and no group is more severely affected prior to the 20122013 Australian Aboriginal
than Australias Indigenous population which and Torres Strait Islander Health Survey [16].
154 S. Colagiuri





2534(b) 3544 4554 55 and over
Age group (years)

Aboriginal and Torres Strait Islander Non-indigenous

Fig. 9.1 Diabetes or high blood sugar levels by Indigenous status and age (Sources: 20122013 Australian Aboriginal
and Torres Strait Islander Health Survey and 20112012 Australian Health Survey)

Among 24 studies, sample sizes varied from 152 sharp rise in diabetes prevalence with age with
to 29,687 participants, with 71 % of studies hav- 31.7 % of those aged 35 years and over and
ing a sample size over 500 people. The mean 52.4 % of those 55 years and over having diabe-
sample study age ranged from 21 to 51 years. tes. Of the people with diabetes, 48 (28.7 %) were
Eighteen studies included blood testing with 14 newly diagnosed of whom 24 (50 %) would not
using WHO diagnostic criteria and 4 the ADA have been diagnosed without an OGTT.
criteria, while 25 % presented self-reported data The Australian Health Survey reported that
and 13 % used medical records. The prevalence 4.7 % of the Indigenous population had impaired
estimates of diabetes ranged from 3.5 % to fasting glucose (IFG) indicating an increased
33.1 %. Even when only considering the 16 stud- risk for future diabetes. After taking age dif-
ies with a mean sample age of 3040 years, the ferences into account, Indigenous people were
diabetes prevalence still varied between 6.5 % nearly twice as likely to be at high risk of diabe-
and 26.2 %. Consequently, the authors were tes compared with non-Indigenous Australians.
unable to establish a true single prevalence esti- The prevalence of IFG increased with age with
mate due to the marked heterogeneity of the data. 7.5 % of those aged 55 and over having IFG [3].
These differences may have been the result of The review by Minges et al. included five studies
major differences in the Indigenous populations, which reported an IGT prevalence of 4.721.1 %
the Indigenous communities, or in the way stud- [16]. With the exception of one study, there was a
ies were conducted. greater prevalence of diabetes than IGT. IFG was
The most recent study which included 2 h only assessed in one study of an island population
OGTT testing was the DRUID (Diabetes and in the top end of Australia and reported a preva-
Related conditions in Urban Indigenous people lence of 5.1 %. The DRUID study [17] reported a
in the Darwin region) study [17] which included prevalence for IGT of 12 %. In contrast to diabe-
861 people aged 15 years or older. The overall tes, both IGT and IFG remained relatively stable
prevalence of diabetes was 19.4 %. There was a in the age groups above 25.
9 Diabetes in Indigenous Australians and Other Underserved Communities in Australia 155

Type 2 Diabetes in Young People were compared with 470 type 1 diabetes subjects
Craig et al. determined the incidence of type with similar age of onset. The median observation
2 diabetes in young people <19 years old and period was similar at 21.4 (interquartile range,
their characteristics in the Indigenous group 1430.7) and 23.4 (15.732.4) years for the type
in a prospective population-based study. From 2 and the type 1 diabetes cohorts, respectively. A
2001 to 2006, there were 128 incident cases of significant mortality excess was noted in the type
type 2 diabetes (62 boys, 66 girls). The median 2 diabetes subjects compared with the type 1 dia-
age at diagnosis was 14.5 years (interquartile betes subjects (11 vs. 6.8 %, P = 0.03), with an
range, 13.016.4), and 90 % were overweight increased hazard for death (hazard ratio, 2.0 [95 %
or obese (BMI >85th percentile for age). Mean CI 1.23.2], P = 0.003). Death in the type 2 diabe-
annual incidence was 2.5/100,000 person-years tes subjects occurred after a significantly shorter
(95 % CI, 2.13.0) in 1018-year-olds. Of the disease duration (26.9 [18.136.0] vs. 36.5 [24.4
ethnic groups represented, white Australian 45.4] years, P = 0.01) and at a relatively young age,
comprised 29 %, Indigenous 22 %, Asian 22 %, and there were more cardiovascular deaths (50 vs.
North African/Middle Eastern 12 %, and Maori/ 30 %, P < 0.05). Despite equivalent glycemic con-
Polynesian/Melanesian 10 %. The incidence trol and shorter disease duration, the prevalence of
of type 2 diabetes was significantly higher in albuminuria and less favorable cardiovascular risk
the Indigenous compared with the non-Indige- factors was greater in the type 2 diabetes subjects
nous group (incidence rate ratio, 6.1; 95 % CI, and neuropathy scores, and macrovascular com-
3.99.7; P < 0.001), but incidence rates of type plications were also increased [21].
1 diabetes were similar (15.5 vs. 21.4/100,000,
respectively). Type 2 diabetes accounted for Risk Factors for Diabetes
11 % of incident cases of diabetes in 1018-year-
olds [18]. Obesity
Type 2 diabetes among Indigenous children The National Aboriginal and Torres Strait
and adolescents appears to be increasing in inci- Islander Health Survey showed that four in every
dence, and the burden is much greater than that ten (39.8 %) Indigenous persons were obese, and
experienced by non-Indigenous young people these obese individuals were around seven times
[19]. Indigenous children and adolescents with more likely than normal weight or underweight
type 2 diabetes typically have a family history of individuals to have diabetes (17.2 % compared
type 2 diabetes and are overweight or obese and with 2.4 %) [15].
may have signs of hyperinsulinism such as acan- Daniel et al. reported the risk of having IGT
thosis nigricans [20]. Onset of type 2 diabetes is and diabetes relative to body mass index (BMI)
usually during early adolescence, and patients are among 2626 aboriginal men and women aged
often asymptomatic at presentation. Data on 1594 years. The population was divided into
comorbidities at diagnosis are lacking and may five strata of BMI (<22, 2224.9, 2529.9,
be a reflection of poor screening. 3034.9, and 35 kg/m2). The prevalence of IGT
There is evidence that young-onset type 2 dia- and diabetes, respectively, adjusted for age and
betes is a more lethal phenotype of diabetes and BMI, was 13.9 % and 14.2 % among men and
is associated with a greater mortality, more diabe- 15.7 % and 15.2 % among women. Odds ratios
tes complications, and unfavorable cardiovascular (95 % CI) for IGT and diabetes for increasing
disease risk factors when compared with type 1 BMI strata 22 kg/m2 ranged from 1.7 (1.02.9)
diabetes. Long-term clinical outcomes and sur- to 5.1 (2.410.5) for IGT and from 2.0 (1.23.5)
vival in young-onset type 2 diabetes were com- to 6.1 (3.311.1) for diabetes. For IGT and diabe-
pared with type 1 diabetes with a similar age of tes, across genders, the population-attributable
onset using an ambulatory diabetes center data- risk percentages (95 % CI) for BMI 22 kg/m2
base. Outcomes in 354 people with type 2 dia- were 34.1 % (26.241.9 %) for IGT and 46.4 %
betes with age of onset between 15 and 30 years (38.554.5 %) for diabetes [22].
156 S. Colagiuri

The DRUID study reported the association Busfield et al. studied an Indigenous Australian
between various indices of obesity and diabetes community with a phenotype characterized by
after adjusting for age. Among males, the odds severe insulin resistance. A genome-wide scan
ratios for diabetes in the highest quartile com- for type 2 diabetes susceptibility genes in a large
pared to the lowest quartile were 10.4 (95 % CI multi-generation pedigree from this community
1.290.6) for WHR, 6.0 (1.229.6) for waist cir- identified a region of significant positive linkage
cumference, and 2.0 (0.66.3) for BMI. Among with type 2 diabetes on chromosome 2q with sev-
females, the corresponding odds ratios were eral candidate genes identified in the region [25].
WHR 25.9 (6.0112.0), waist circumference 7.5 Another genome-wide analysis found peaks on
(2.919.2), and BMI 4.8 (2.110.8) [17]. chromosomes 1 and 21. On chromosome 1, the
gene had a strong association with type 2 diabe-
Low Birth Weight tes, hyperlipidemia, and blood pressure in type
Child health, and especially low birth weight, is 2 diabetes in European populations. The gene
considered an important factor in the develop- with a strong association with type 2 diabetes on
ment of diabetes. In 2011, the low birth weight chromosome 21 is involved with regulating insu-
rate for babies born to Indigenous mothers was lin [26]. Further studies are required to replicate
twice the rate for those with a non-Indigenous these findings in other Indigenous and interna-
mother (13 % compared with 6 %) and was sig- tional populations.
nificantly higher in remote areas (15 %) than in
non-remote areas (12 %) [23]. Despite these high Gestational Diabetes
rates, there has been a significant 9 % decline in Ishak and Petocz investigated the prevalence,
low birth weight rates between 2000 and 2011 in trends, and risk factors of gestational diabetes
Indigenous mothers. mellitus (GDM) in Indigenous Australians com-
A major factor in low birth weight is smoking. pared with the non-Indigenous population in a
Fifty percent of Indigenous women smoked dur- retrospective population analysis of 230,011
ing pregnancy in 2011, four times the rate for deliveries from an administrative database in the
non-Indigenous women, and 51 % of low birth state of South Australia between 1988 and 1999.
weight births to Indigenous mothers has been The age-standardized GDM rate for Indigenous
attributed to smoking, compared with 19 % for mothers was more than 2.5 times higher than
non-Indigenous mothers. After adjusting for age that for non-Indigenous mothers (4.3 vs. 1.8 %)
differences and other factors, it is estimated that with no significant trend changes over the time
if the smoking rate for Indigenous pregnant period of the study [27]. However, a more recent
women was the same as for other Australian systematic review of diabetes in pregnancy
mothers, the proportion of low birth weight included 11 studies which reported prevalence of
babies could be reduced by 26 % [14]. GDM in Indigenous Australians. The overall rate
of GDM was higher at 8.4 % compared with
Social Determinants 25 % worldwide [28].
There remains a significant gap between
Indigenous and non-Indigenous Australians in
several parameters which can impact health. These People Living in Rural and Remote
include education attainment rates, employment, Areas
household incomes (43 % of Indigenous adults
were in the lowest quintile compared with 17 % of Australias population is largely concentrated in
non-Indigenous adults), and homelessness [14]. the east and southeast of the country, and most
Australians live in capital cities with the majority
Genetic Studies living in major cities (71 %), 18 % in inner
There have been a limited number of gene- regional areas, 9 % in outer regional areas, 1.4%
related studies including studies on diabetes sus- in remote areas, and 1% in very remote areas.
ceptibility genes in Indigenous Australians [24]. The proportion living in major cities has increased
9 Diabetes in Indigenous Australians and Other Underserved Communities in Australia 157

over the past decade, while the population in very remote areas are between 10 % and 70 % as high
remote areas has fallen [29]. as in major cities. Cardiovascular diseases are
Health outcomes, such as higher rates of death, responsible for nearly a third of the elevated male
are worse outside of major cities related to differ- death rates outside major cities. Male death rates
ences in access to services, risk factors, and the from diabetes are 1.3 times as high in inner
remote environment, although it is not possible to regional areas and 3.7 times as high in very
apportion the generally poorer health outcomes remote areas compared with major cities [31].
outside major cities to these factors [30]. With In 2010, 5.4 % of all deaths that year were
respect to health services, there are lower rates of attributed to diabetes and causes related to diabe-
some hospital surgical procedures, lower rates of tes, although as is well known this will be an
primary care physician consultation, and gener- underestimate. Diabetes-related death rate was
ally higher rates of hospital admission in regional 55 % higher among males than females (39
and remote areas than in major cities. There are deaths per 100,000 males and 25 deaths per
also significant differences in interregional health 100,000 females) [32]. Death rate from diabetes-
behavior and risk factors. People in regional and related causes increased with increasing remote-
remote areas are more likely than their urban ness 28 deaths per 100,000 population among
counterparts to be a daily smoker (outer regional people living in major cities, 32 deaths per
and remote 22 % and inner regional 18 % com- 100,000 population among people living in inner
pared with 15 % in major cities), be overweight or regional areas, and 43 deaths per 100,000 popula-
obese (70 % and 69 % compared with 60 %), be tion among people living in outer regional,
insufficiently active (60 % and 63 % compared remote, and very remote areas.
with 54 %), drink alcohol in harmful quantities Diabetes prevalence is also linked with region
(24 % and 21 % compared with 19 %), and have of residence. Overall in Australia, the age-
high blood cholesterol (37 % and 38 % compared standardized prevalence of diabetes in 2011
with 31 %) [31]. In addition, over half of outer 2012 was 4.2 %. Rates of diabetes were 3.9 % in
regional, remote, and very remote residents live in major cities and 4.9 % in outer regional and
areas classified as the lowest socioeconomic sta- remote areas [3].
tus compared with around one-quarter of people The situation is worse for Indigenous people
in major cities and 77 % in very remote areas. living in remote areas with around one in five
Another contributing factor to the health sta- (20.8 %) having diabetes compared with around
tus of people living in rural and remote areas is one in ten people in non-remote areas (9.4 %).
the higher proportion of Indigenous people and This difference is particularly pronounced for
their generally poorer health outcomes. Most newly diagnosed diabetes, which was five times
Indigenous Australians live in urban areas 35 % as high in remote areas than in non-remote areas
in major cities, 22 % in each of inner and outer (4.8 % compared with 0.9 %) [3]. In addition,
regional areas, and the remaining 21 % in either Indigenous adults in remote areas are less likely
remote or very remote areas. Nevertheless, they to have their diabetes effectively managed
make up a relatively large proportion of the popu- (25.1 % compared with 43.5 %) and two and a
lation living in remote areas of Australia 45 % half times more likely to have chronic kidney dis-
of all people living in very remote areas and 16 % ease (33.6 % compared with 13.1 %).
living in remote areas [30].
Overall, death rates increase with increasing
remoteness. In 2012, the age-standardized rate People from Culturally
was highest in very remote areas (840 per 100,000 and Linguistically Diverse
population), followed by remote (670 per 100,000 Backgrounds
population), outer regional (640 per 100,000
population), inner regional (610 per 100,000 Around 28 % of Australias population was born
population), and major cities (550 per 100,000 overseas with migration from more than 200
population) [31]. Death rates in regional and countries around the world [33]. The proportion
158 S. Colagiuri

of the population born overseas has grown 50 years. For both men and women, odds of type
steadily from one in ten (10 %) in 1947. Between 2 diabetes were higher for all migrant groups
2004 and 2014, Australias overseas-born popu- than the Australian-born reference population.
lation increased from 4.8 million to 6.6 million After adjusting for age and SES, odds were
people. The largest proportion was born in the 6.3 and 7.2 times higher for men and women
United Kingdom (5.2 %) followed by New born in the Pacific Islands, respectively, and
Zealand (2.6 %), China (excluding Hong Kong) 5.2 and 5.0 times higher for men and women
(1.9 %), India (1.7 %), the Philippines (1 %), born in Southern and Central Asia, respec-
Vietnam (1.0 %), Italy (0.9 %), South Africa tively. However, compared with Australian-born
(0.8 %), Malaysia (0.7 %), and Germany (0.5 %) people, age- and SES-adjusted prevalence odds
[33]. The relative proportion of Australian resi- were significantly increased across all groups for
dents born overseas is changing with a decrease males and females including Oceania (2.6 and
in those born in the United Kingdom and an 3.0), Northwest Europe (1.5 and 1.7), Southern
increase in those born in New Zealand, China, and Eastern Europe (2.0 and 2.4), North Africa
and India. Of overseas-born Australians, the big- and the Middle East (4.0 and 4.7), Southeast
gest proportion come from Northwest Europe Asia (3.1 and 4.0), Northeast Asia (1.9 and 2.6),
(including the United Kingdom) (25 %) followed Southern and Central Asia (5.2 and 5.0), America
by Southeast Asia (14 %), Southern and Eastern (2.0 and 2.4), and sub-Saharan Africa (2.7 and
Europe (12 %), Oceania (including New Zealand) 3.1). These sociocultural differences have impli-
(12 %), Northeast Asia (12 %), and Southern and cations for health service planning and delivery,
Central Asia (10 %). This increase in cultural policy, and preventive efforts in Australia.
diversity has implications for health services and Gestational diabetes is a particular problem in
the demographics of health in Australia [30]. mothers from different ethnic groups. A computer-
Culturally and linguistically diverse (CALD) ized database of all births (n = 956,738) between
communities in Australia experience both signifi- 1995 and 2005 in the Australian state of New South
cant health disparities and a lack of access to ser- Wales was used to examine the association between
vices. A systematic review of the literature on the sociodemographic characteristics and the occur-
effectiveness of culturally appropriate interven- rence of gestational diabetes [36]. Between 1995
tions to manage or prevent chronic disease in and 2005, the prevalence of GDM increased by
CALD communities found that the health of 45 %, from 3.0 % to 4.4 %. Women born in South
Australias CALD population is poor in compari- Asia had the highest adjusted odds ratio of any
son to the general population. Hospital admis- region (4.22 [95 % CI 4.014.44]) relative to
sions for CALD people were more than double, women born in Australia. However, all regions
particularly for chronic and disabling conditions, showed increased adjusted odds Northeast and
such as diabetes, traumatic injury, heart and kid- Southeast Asia 3.24 (3.163.34), Europe and North
ney disease, and respiratory problems [34]. America 1.21 (1.161.26), Middle East and North
The prevalence of type 2 diabetes varies by Africa 2.40 (2.302.51), Pacific 2.94 (2.783.11),
ethnicity and socioeconomic status (SES), and other Africa 1.62 (1.461.80), and Caribbean,
migrants experience a disproportionate burden Central, and South America 1.82 (1.652.01).
of disease compared with locally born groups. The rate of progression from GDM to devel-
Abouzeid et al. compared the prevalence odds of oping postpartum abnormal glucose tolerance is
type 2 diabetes among immigrant groups in the greater in certain ethnic groups. In a prospective
Australian state of Victoria with Australian-born study of 101 women who had GDM, ethnicity
residents. The overall prevalence of diagnosed was a major risk factor for the development of
type 2 diabetes in Victoria was 4.1 % in men diabetes during a mean follow-up of 5.5 years
and 3.5 % in women [35]. Of those with type 2 [37]. South Asian women had a significantly
diabetes, over one in five born in Oceania and higher risk of developing abnormal glucose toler-
in Southern and Central Asia were aged under ance (69 %) than women of all other ethnicities.
9 Diabetes in Indigenous Australians and Other Underserved Communities in Australia 159

The prevalence of diabetes and impaired glucose Albumin/creatinine ratio less than 3.5 mg/mmol
tolerance was also very high among other for women and less than 2.5 mg/mmol for men
groups Southeast and East Asian (41 %), Middle Urinary albumin excretion less than 20 mg/L
Eastern (44 %), South European backgrounds Blood pressure less than or equal to
(42 %), and Australian-born women (39 %). 130/80 mmHg
Differences in the clinical characteristics and Normal body mass index (i.e., a BMI score
outcomes of women with GDM from various eth- of between 18.5 and 24.9)
nic groups have also been reported. Wong reported Nonsmoker
a retrospective review of 827 women with GDM
from five ethnic groups (Southeast Asian, South In 20122013, around two in five (38.9 %)
Asian, Middle Eastern, Anglo-European, and Indigenous Australian adults with known diabetes
Pacific Islander). Southeast Asians had the lowest had an HbA1c result of 7.0 % or less. Overall,
BMI and lowest need for insulin therapy and their Indigenous women were more likely than men to
offspring had the lowest rate of macrosomia and achieve this result (47.0 % compared with 28.1 %).
lowest birth weight. In contrast, women from Indigenous people with known diabetes were less
Pacific Islands had the highest BMI and greatest likely than their non-Indigenous counterparts to
need for insulin therapy, and their offspring had have an HbA1c of 7.0 % or less (38.9 % compared
the highest birth weights. This study highlighted with 55.9 %) [15].
the significant differences in clinical characteris- Just over half (56.9 %) met the management
tics of women with GDM [38]. target for triglycerides and 44.4 % met the target
for albumin/creatinine ratio (ACR). However,
Indigenous people with diabetes were less likely
Diabetes Management than non-Indigenous people to meet these targets,
particularly for ACR (44.4 % compared with
In general, information on diabetes management 71.0 %) [15].
and rates of complications in Australia are less Two of the most common problems in
well documented than epidemiological data. Indigenous people are cardiovascular (CVD) and
While there are some data for Indigenous kidney disease. CVD risk factors are prevalent.
Australians, there are few specific data for the The DRUID study found that while people with
other two groups included in this review. diabetes had a greater number of CVD risk fac-
Indigenous Australians have evidence of tors than those without diabetes, these risk fac-
poorer glycemic and metabolic control and high tors were relatively common in young people
rates of diabetes complications and cardiovascu- (aged <35 years) without diabetes with almost
lar mortality. half (45 %) having at least two risk factors and
In the National Aboriginal and Torres Strait only 18 % having none [17]. Six in ten (60.5 %)
Islander Health Measures Survey, the following Indigenous people with diabetes had lower than
goals on the optimal management in those with normal levels of HDL cholesterol compared with
known diabetes were assessed: 32.9 % of those without diabetes (Fig. 9.2). They
were also around twice as likely to have high tri-
Fasting blood glucose between 6.0 and glycerides (45.1 % compared with 20.7 %). A
8.0 mmol/L quarter (25.0 %) had high cholesterol, but only
HbA1c levels less than or equal to 7.0 % around one in ten (9.1 %) of this group were
Total cholesterol less than 4.0 mmol/L aware they had it [15]. Many Indigenous people
HDL cholesterol greater than or equal to with diabetes also had signs of other chronic con-
1.0 mmol/L ditions. Around half (53.1 %) had signs of chronic
LDL cholesterol less than 2.0 mmol/L kidney disease, significantly higher than the cor-
Non-HDL cholesterol less than 2.5 mmol/L responding rate in the non-Indigenous population
Triglycerides less than 2.0 mmol/L (32.5 %) [15].
160 S. Colagiuri

Fig. 9.2 Selected chronic %

disease biomarkers in
Indigenous subjects with and 60
without diabetes (Source:
20122013 Australian
Aboriginal and Torres Strait 50
Islander Health Survey
Biomedical Results)




Abnormal HDL good Abnormal Abnormal GGT Has anaemia
cholesterol triglycerides (liver test)

Chronic disease biomarkers

Does not have diabetes Has diabetes

Thomas et al. examined the management of between Indigenous Australians and Anglo-Celt
144 Indigenous people with type 2 diabetes in patients in an urban Australian community [40].
Australian primary care compared with that in Data from the Fremantle Diabetes Study col-
non-Indigenous patients presenting consecutively lected from 1993 to 1996 (phase I) and from
to the same medical practitioner (n = 449). 2008 to 2011 (phase II) were analyzed. Indigenous
Indigenous Australians had high rates of micro- participants were younger at entry and, at diabe-
and macrovascular disease with 60 % having an tes diagnosis, were less likely to be educated
abnormal ACR compared to 33 % of non- beyond primary level and more likely to be smok-
Indigenous patients (p < 0.01). In addition, they ers. HbA1c decreased in both groups over time
were more likely to have established macrovascu- (Indigenous median 9.6 % [interquartile range
lar disease (adjusted odds ratio 2.7). This excess in 7.810.7 %] to 8.4 % [6.610.6 %] vs. Anglo-Celt
complications was associated with poor glycemic median 7.1 % [6.28.4 %] to 6.7 % [6.27.5 %]),
control, with an HbA1c 8.0 %, observed in 55 % but the gap persisted. Indigenous patients were
of all Indigenous patients, despite the similar fre- more likely to have microvascular disease in both
quency of use of oral antidiabetic agents and insu- phases. The prevalence of peripheral arterial dis-
lin. Smoking was more common in Indigenous ease (ankle-brachial index <0.90 or lower-
patients (38 % vs. 10 %, p < 0.01). However, the extremity amputation) increased in Indigenous
achievement of LDL and blood pressure targets but decreased in Anglo-Celt participants.
was the same or better in Indigenous patients. This While renal disease is common in Indigenous
cross-sectional study confirmed aboriginal ethnic- Australians, there is considerable regional varia-
ity as a powerful risk factor for microvascular and tion in the incidence of end-stage renal disease
macrovascular disease [39]. (ESRD). Cass and colleagues reviewed informa-
Davis et al. used data from the observational tion on 719 Indigenous ESRD patients who com-
Fremantle Diabetes Study to examine disparities menced treatment in Australia during 19931998
in the nature and management of type 2 diabetes using data from the Australian and New Zealand
9 Diabetes in Indigenous Australians and Other Underserved Communities in Australia 161

Dialysis and Transplant Registry who started 2 years of treatment. In these 118, blood pressures
ESRD treatment. Standardized ESRD incidence fell significantly, while ACR and GFR stabilized.
among Indigenous Australians was highest in Rates of the combined end points of renal failure
remote regions, where it was up to 30 times the and natural death per 100 person-years were 2.9
national incidence for all Australians. In urban for the treatment group (95 % CI, 1.74.6) and
regions, the standardized incidence was much 4.8 for the historical control group (95 % CI, 3.3
lower, but remained significantly higher than the 7.0). After adjustment for baseline ACR category,
national incidence. Forty-eight percent of the relative risk of the treatment group versus the
Indigenous ESRD patients came from regions control group for these combined end points was
without dialysis or transplant facilities and 16.3 % 0.47 (95 % CI, 0.250.86; P = 0.013). Treatment
from regions with only satellite dialysis facilities. benefit was especially marked in people with
Because of the location of treatment centers, overt albuminuria or hypertension and in people
there is inequitable access to ESRD treatment without diabetes. The estimates of benefit were
services for a significant proportion of Indigenous supported by a fall in community rates of death
patients [41]. and renal failure.
A number of attempts have been made to Point of care testing (POCT) for HbA1c and
improve clinical systems for addressing the gap urine ACR is increasingly available in Indigenous
in care of Indigenous Australians with diabetes. communities under the auspices of the Quality
McDermott and colleagues assessed changes in Assurance for Aboriginal Medical Services
clinical indicators of adults diagnosed with dia- (QAAMS) Program. Shephard et al. evaluated
betes using audits of clinical records of Torres the QAAMS Program and assessed satisfaction
Strait Islander adults on diabetes registers in 21 with POCT for HbA1c and urine ACR in a medi-
primary care clinics. Over a 6-year period, the cal service in a remote area of Northern Australia
number of adults included on the diabetes regis- for Indigenous people with type 2 diabetes [44].
ter increased from 555 in 1999 to 1024 in 2005. Both doctors and patients reported that the imme-
Mean weight increased from 86.8 to 95.6 kg, and diacy of POCT had contributed positively to the
mean HbA1c level remained unchanged at about identification and management of diabetes,
9 %, but the proportion with HbA1c level <7 % improved doctor-patient relationship, and facili-
increased from 18.4 % to 26.1 % and the propor- tated compliance and self-motivation to control
tion with BP <140/90 mmHg increased from diabetes. At the end of the 12 months, there was a
40.3 % to 66.8 %. This study showed that some statistically significant drop in HbA1c from 9.3 %
improvements can be achieved by introducing to 8.6 % (P = 0.003) with an improvement in the
clinical systems, but important clinical parame- percentage of patients controlling their diabetes.
ters such as weight gain and hyperglycemia The addition of HbA1c as a diagnostic cri-
remain a challenge [42]. terion for diabetes has simplified algorithms
The effectiveness of a systematic treatment for testing and detecting undiagnosed diabetes.
program to modify renal and cardiovascular dis- POCT for HbA1c in this context has particular
ease in an Indigenous Australian community advantages in Indigenous communities. Marley
from the Tiwi Islands with high rates of renal and colleagues compared the Australian glucose-
failure and cardiovascular deaths was reported by based algorithm [45] with an HbA1c-based algo-
Hoy et al. [43]. Adults with blood pressure rithm applied in a remote Australian Aboriginal
140/90, with diabetes and ACR 3.4 g/mol, or community. The HbA1c-based algorithm used
with progressive overt albuminuria were offered an initial POCT HbA1c assessment followed by
the intervention which included attempts to laboratory HbA1c assay if needed. Participants
achieve blood pressure goals and improve control were significantly more likely to receive a defini-
of blood glucose and lipid levels and health edu- tive result within 7 days and to be diagnosed
cation. Two hundred fifty-eight people enrolled with diabetes using the HbA1c algorithm than
in the program, and 118 had complete data for with the glucose-based protocol. The study also
162 S. Colagiuri

highlighted the increased likelihood of follow-up 4. Dunstan DW, Zimmet PZ, Welborn TA, et al. The ris-
ing prevalence of diabetes and impaired glucose toler-
with HbA1c testing; only 42 % of participants
ance: the Australian diabetes, obesity and lifestyle
with an equivocal glucose result underwent an study. Diabetes Care. 2002;25:82934.
OGTT as recommended by the Australian guide- 5. Australia Institute of Health and Welfare Australias
line [46]. Health. Australias health series no 12. Cat no. AUS
122. Canberra: AIHW; 2010.
6. Australian Institute of Health and Welfare (AIHW).
Conclusions Causes of death. 2013. [Online]. Available: http://
Like other countries around the world, Australia
is experiencing a steady increase in the number 7. Australian Institute of Health and Welfare (AIHW).
Diabetes: Australian facts 2008. Canberra: Diabetes
of people with diabetes which is placing a con-
Series; 2008.
siderable burden on individuals, their families, 8. Colagiuri S, Brnabic A, Gomez M, et al. DiabCo$t
and the whole society. The impact on Australia: assessing the burden of type 1 diabetes in
Indigenous Australians is particularly heavy Australia. Canberra: Diabetes Australia; 2009.
9. Lee CM, Colagiuri R, Magliano DJ, et al. The cost of
with diabetes prevalence rates, threefold higher
diabetes in Adults in Australia. Diabetes Res Clin
than non-Indigenous Australians. Indigenous Pract. 2013;99(3):385904.
Australians also have increased rates of compli- 10. Australian Bureau of Statistics.
cations, especially renal. Two other groups also ausstats/abs@.nsf/Lookup/by%20Subject/4338.0~2011-
experience significant disadvantage people
from culturally and linguistically diverse back- 11. ABS. Projected population, Aboriginal and Torres
grounds and people living in rural and remote Strait Islander Australians, Australia, states and terri-
areas. While epidemiological data are available tories, 20112026. ABS cat. no. 3238.0. Canberra:
ABS; 2014.
for these groups, information on diabetes man-
12. ABS. Life tables for Aboriginal and Torres Strait
agement and rates of complications are less Islander Australians, 20102012. ABS cat. no.
well documented. While some programs are 3302.0.55.003. Canberra: ABS; 2013.
being implemented to improve quality of care 13. AIHW. Mortality and life expectancy of Indigenous
Australians 2008 to 2012. Cat. no. IHW 140.
and outcomes, an increased effort is needed to
Canberra: AIHW; 2014.
reduce the care gap in these groups compared 14. AIHW. Aboriginal and Torres Strait Islander health
with other Australians. performance framework 2014 report: detailed analy-
There are many gaps in our knowledge of ses. 2015. Cat. no. IHW 167.
diabetes in these populations and unmet needs
15. ABS. Australian Aboriginal and Torres Strait Islander
with respect to prevention, care, and treat- Health Survey: biomedical results, 2012-13. 2014.
ment. Future research will be needed to Cat. no. 4727.0.55.003.
address these gaps and reduce the disparity 16. Minges KE, Zimmet P, Magliano DJ, et al. Diabetes
prevalence and determinants in Indigenous Australian
between these groups and other Australians.
populations: a systematic review. Diabetes Res Clin
Pract. 2011;93(2):13949.
17. ODea K, Cunningham J, Maple-Brown L,
Weeramanthri T, Shaw J, Dunbar T, et al. Diabetes
and cardiovascular risk factors in urban Indigenous
References adults: results from the DRUID study. Diabetes Res
Clin Pract. 2008;80:4839.
1. Australian National Diabetes Strategy 2016-2020. 18. Craig ME, Femia G, Broyda V, et al. Type 2 diabetes in Indigenous and non-Indigenous children and ado-
nsf/Content/nds-2016-2020. lescents in New South Wales. Med J Aust. 2007;186:
2. National Diabetes Service Scheme (NDSS). Data 4979.
snapshots. 2014. [Online]. Available: http://www. 19. Azzopardi P, Brown AD, Zimmet P, Fahy RE, Dent GA, Kelly MJ, Kranzusch K, Maple-Brown LJ,
3. Australian Bureau of Statistics (ABS). Australian Nossar V, Silink M, Sinha AK, Stone ML. Type 2 d